<!DOCTYPE article PUBLIC "-//NLM//DTD JATS (Z39.96) Journal Archiving and Interchange DTD with MathML3 v1.2 20190208//EN" "JATS-archivearticle1-mathml3.dtd"> <article xmlns:ali="http://www.niso.org/schemas/ali/1.0/" xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink" article-type="research-article" dtd-version="1.2"><front><journal-meta><journal-id journal-id-type="nlm-ta">elife</journal-id><journal-id journal-id-type="publisher-id">eLife</journal-id><journal-title-group><journal-title>eLife</journal-title></journal-title-group><issn pub-type="epub" publication-format="electronic">2050-084X</issn><publisher><publisher-name>eLife Sciences Publications, Ltd</publisher-name></publisher></journal-meta><article-meta><article-id pub-id-type="publisher-id">69799</article-id><article-id pub-id-type="doi">10.7554/eLife.69799</article-id><article-categories><subj-group subj-group-type="display-channel"><subject>Research Article</subject></subj-group><subj-group subj-group-type="heading"><subject>Cancer Biology</subject></subj-group><subj-group subj-group-type="heading"><subject>Computational and Systems Biology</subject></subj-group></article-categories><title-group><article-title>Quantifying chromosomal instability from intratumoral karyotype diversity using agent-based modeling and Bayesian inference</article-title></title-group><contrib-group><contrib contrib-type="author" id="author-236093"><name><surname>Lynch</surname><given-names>Andrew R</given-names></name><contrib-id authenticated="true" contrib-id-type="orcid">https://orcid.org/0000-0002-0238-682X</contrib-id><xref ref-type="aff" rid="aff1">1</xref><xref ref-type="aff" rid="aff2">2</xref><xref ref-type="other" rid="fund4"/><xref ref-type="other" rid="fund5"/><xref ref-type="other" rid="fund6"/><xref ref-type="fn" rid="con1"/><xref ref-type="fn" rid="conf1"/></contrib><contrib contrib-type="author" id="author-203982"><name><surname>Arp</surname><given-names>Nicholas L</given-names></name><contrib-id authenticated="true" contrib-id-type="orcid">https://orcid.org/0000-0001-8709-0667</contrib-id><xref ref-type="aff" rid="aff1">1</xref><xref ref-type="other" rid="fund7"/><xref ref-type="other" rid="fund9"/><xref ref-type="fn" rid="con2"/><xref ref-type="fn" rid="conf2"/></contrib><contrib contrib-type="author" id="author-236365"><name><surname>Zhou</surname><given-names>Amber S</given-names></name><xref ref-type="aff" rid="aff1">1</xref><xref ref-type="aff" rid="aff2">2</xref><xref ref-type="other" rid="fund8"/><xref ref-type="fn" rid="con3"/><xref ref-type="fn" rid="conf1"/></contrib><contrib contrib-type="author" id="author-132941"><name><surname>Weaver</surname><given-names>Beth A</given-names></name><contrib-id authenticated="true" contrib-id-type="orcid">https://orcid.org/0000-0002-7830-3816</contrib-id><xref ref-type="aff" rid="aff1">1</xref><xref ref-type="aff" rid="aff2">2</xref><xref ref-type="aff" rid="aff3">3</xref><xref ref-type="fn" rid="con4"/><xref ref-type="fn" rid="conf2"/></contrib><contrib contrib-type="author" corresp="yes" id="author-89000"><name><surname>Burkard</surname><given-names>Mark E</given-names></name><contrib-id authenticated="true" contrib-id-type="orcid">https://orcid.org/0000-0002-4215-7722</contrib-id><email>mburkard@wisc.edu</email><xref ref-type="aff" rid="aff1">1</xref><xref ref-type="aff" rid="aff2">2</xref><xref ref-type="aff" rid="aff4">4</xref><xref ref-type="other" rid="fund1"/><xref ref-type="other" rid="fund2"/><xref ref-type="other" rid="fund3"/><xref ref-type="fn" rid="con5"/><xref ref-type="fn" rid="conf3"/></contrib><aff id="aff1"><label>1</label><institution-wrap><institution-id institution-id-type="ror">https://ror.org/01y2jtd41</institution-id><institution>Carbone Cancer Center, University of Wisconsin-Madison</institution></institution-wrap><addr-line><named-content content-type="city">Madison</named-content></addr-line><country>United States</country></aff><aff id="aff2"><label>2</label><institution-wrap><institution-id institution-id-type="ror">https://ror.org/01y2jtd41</institution-id><institution>McArdle Laboratory for Cancer Research, University of Wisconsin-Madison</institution></institution-wrap><addr-line><named-content content-type="city">Madison</named-content></addr-line><country>United States</country></aff><aff id="aff3"><label>3</label><institution-wrap><institution-id institution-id-type="ror">https://ror.org/01y2jtd41</institution-id><institution>Department of Cell and Regenerative Biology, University of Wisconsin</institution></institution-wrap><addr-line><named-content content-type="city">Madison</named-content></addr-line><country>United States</country></aff><aff id="aff4"><label>4</label><institution-wrap><institution-id institution-id-type="ror">https://ror.org/01y2jtd41</institution-id><institution>Division of Hematology Medical Oncology and Palliative Care, Department of Medicine University of Wisconsin</institution></institution-wrap><addr-line><named-content content-type="city">Madison</named-content></addr-line><country>United States</country></aff></contrib-group><contrib-group content-type="section"><contrib contrib-type="editor"><name><surname>Marston</surname><given-names>Adèle L</given-names></name><role>Reviewing Editor</role><aff><institution-wrap><institution-id institution-id-type="ror">https://ror.org/01nrxwf90</institution-id><institution>University of Edinburgh</institution></institution-wrap><country>United Kingdom</country></aff></contrib><contrib contrib-type="senior_editor"><name><surname>Akhmanova</surname><given-names>Anna</given-names></name><role>Senior Editor</role><aff><institution-wrap><institution-id institution-id-type="ror">https://ror.org/04pp8hn57</institution-id><institution>Utrecht University</institution></institution-wrap><country>Netherlands</country></aff></contrib></contrib-group><pub-date date-type="publication" publication-format="electronic"><day>05</day><month>04</month><year>2022</year></pub-date><pub-date pub-type="collection"><year>2022</year></pub-date><volume>11</volume><elocation-id>e69799</elocation-id><history><date date-type="received" iso-8601-date="2021-04-27"><day>27</day><month>04</month><year>2021</year></date><date date-type="accepted" iso-8601-date="2022-04-01"><day>01</day><month>04</month><year>2022</year></date></history><pub-history><event><event-desc>This manuscript was published as a preprint at .</event-desc><date date-type="preprint" iso-8601-date="2021-04-27"><day>27</day><month>04</month><year>2021</year></date><self-uri content-type="preprint" xlink:href="https://doi.org/10.1101/2021.04.26.441466"/></event></pub-history><permissions><copyright-statement>© 2022, Lynch et al</copyright-statement><copyright-year>2022</copyright-year><copyright-holder>Lynch et al</copyright-holder><ali:free_to_read/><license xlink:href="http://creativecommons.org/licenses/by/4.0/"><ali:license_ref>http://creativecommons.org/licenses/by/4.0/</ali:license_ref><license-p>This article is distributed under the terms of the <ext-link ext-link-type="uri" xlink:href="http://creativecommons.org/licenses/by/4.0/">Creative Commons Attribution License</ext-link>, which permits unrestricted use and redistribution provided that the original author and source are credited.</license-p></license></permissions><self-uri content-type="pdf" xlink:href="elife-69799-v2.pdf"/><self-uri content-type="figures-pdf" xlink:href="elife-69799-figures-v2.pdf"/><abstract><p>Chromosomal instability (CIN)—persistent chromosome gain or loss through abnormal mitotic segregation—is a hallmark of cancer that drives aneuploidy. Intrinsic chromosome mis-segregation rate, a measure of CIN, can inform prognosis and is a promising biomarker for response to anti-microtubule agents. However, existing methodologies to measure this rate are labor intensive, indirect, and confounded by selection against aneuploid cells, which reduces observable diversity. We developed a framework to measure CIN, accounting for karyotype selection, using simulations with various levels of CIN and models of selection. To identify the model parameters that best fit karyotype data from single-cell sequencing, we used approximate Bayesian computation to infer mis-segregation rates and karyotype selection. Experimental validation confirmed the extensive chromosome mis-segregation rates caused by the chemotherapy paclitaxel (18.5 ± 0.5/division). Extending this approach to clinical samples revealed that inferred rates fell within direct observations of cancer cell lines. This work provides the necessary framework to quantify CIN in human tumors and develop it as a predictive biomarker.</p></abstract><abstract abstract-type="executive-summary"><title>eLife digest</title><p>DNA contains all the information that cells need to function. The DNA inside cells is housed in structures called chromosomes, and most healthy human cells contain 23 pairs. When a cell divides, all chromosomes are copied so that each new cell gets a complete set. However, sometimes the process of separating chromosomes is faulty, and new cells may get incorrect numbers of chromosomes during cell division. Cancer cells frequently exhibit this behavior, which is called chromosomal instability’, or CIN.</p><p>Chromosomal instability affects many cancer cells with varying severity. In cancers with high chromosomal instability, the number of chromosomes may change almost every time the cells divide. These cancers are often the most aggressive and difficult to treat.</p><p>Scientists can estimate chromosomal instability by counting differences in the number of chromosomes across many cells. However, many cells that are missing chromosomes die, resulting in inaccurate measures of chromosomal instability. To find a solution to this problem, Lynch et al. counted chromosomes in human cells with different levels of chromosomal instability and created a computer model to work out the relationship between chromosomal instability and chromosome number.</p><p>The model could account for both living and dead cells, which gave more accurate results. Lynch et al. then confirmed the accuracy of their approach by using it on a group of cells treated with a chemotherapy drug that causes a known level of chromosomal instability. They also used existing data from breast and bowel cancer, which revealed that levels of chromosomal instability varied between one mistake per three to twenty cell divisions.</p><p>Lower levels of chromosomal instability can be linked to a better prognosis for cancer patients, but it currently cannot be measured reliably. These results may help to reveal the causes of chromosomal instability and the role it has in cancer. If this method is successfully applied to patient samples, it could also improve our ability to predict how each cancer will progress and may lead to better treatments.</p></abstract><kwd-group kwd-group-type="author-keywords"><kwd>aneuploidy</kwd><kwd>single-cell sequencing</kwd><kwd>agent-based modeling</kwd><kwd>mitosis</kwd><kwd>approximate Bayesian computation</kwd></kwd-group><kwd-group kwd-group-type="research-organism"><title>Research organism</title><kwd>Human</kwd></kwd-group><funding-group><award-group id="fund1"><funding-source><institution-wrap><institution-id institution-id-type="FundRef">http://dx.doi.org/10.13039/100000054</institution-id><institution>National Cancer Institute</institution></institution-wrap></funding-source><award-id>R01CA234904</award-id><principal-award-recipient><name><surname>Burkard</surname><given-names>Mark E</given-names></name></principal-award-recipient></award-group><award-group id="fund2"><funding-source><institution-wrap><institution-id institution-id-type="FundRef">http://dx.doi.org/10.13039/100000002</institution-id><institution>National Institutes of Health</institution></institution-wrap></funding-source><award-id>R01GM141068</award-id><principal-award-recipient><name><surname>Burkard</surname><given-names>Mark E</given-names></name></principal-award-recipient></award-group><award-group id="fund3"><funding-source><institution-wrap><institution-id institution-id-type="FundRef">http://dx.doi.org/10.13039/100000054</institution-id><institution>National Cancer Institute</institution></institution-wrap></funding-source><award-id>P30CA014520</award-id><principal-award-recipient><name><surname>Burkard</surname><given-names>Mark E</given-names></name></principal-award-recipient></award-group><award-group id="fund4"><funding-source><institution-wrap><institution-id institution-id-type="FundRef">http://dx.doi.org/10.13039/100000054</institution-id><institution>National Cancer Institute</institution></institution-wrap></funding-source><award-id>F31CA254247</award-id><principal-award-recipient><name><surname>Lynch</surname><given-names>Andrew R</given-names></name></principal-award-recipient></award-group><award-group id="fund5"><funding-source><institution-wrap><institution-id institution-id-type="FundRef">http://dx.doi.org/10.13039/100000002</institution-id><institution>National Institutes of Health</institution></institution-wrap></funding-source><award-id>T32HG002760</award-id><principal-award-recipient><name><surname>Lynch</surname><given-names>Andrew R</given-names></name></principal-award-recipient></award-group><award-group id="fund6"><funding-source><institution-wrap><institution-id institution-id-type="FundRef">http://dx.doi.org/10.13039/100000002</institution-id><institution>National Institutes of Health</institution></institution-wrap></funding-source><award-id>T32GM81061</award-id><principal-award-recipient><name><surname>Lynch</surname><given-names>Andrew R</given-names></name></principal-award-recipient></award-group><award-group id="fund7"><funding-source><institution-wrap><institution-id institution-id-type="FundRef">http://dx.doi.org/10.13039/100000002</institution-id><institution>National Institutes of Health</institution></institution-wrap></funding-source><award-id>T32GM008692</award-id><principal-award-recipient><name><surname>Arp</surname><given-names>Nicholas L</given-names></name></principal-award-recipient></award-group><award-group id="fund8"><funding-source><institution-wrap><institution-id institution-id-type="FundRef">http://dx.doi.org/10.13039/100000002</institution-id><institution>National Institutes of Health</institution></institution-wrap></funding-source><award-id>T32GM008688</award-id><principal-award-recipient><name><surname>Zhou</surname><given-names>Amber S</given-names></name></principal-award-recipient></award-group><award-group id="fund9"><funding-source><institution-wrap><institution-id institution-id-type="FundRef">http://dx.doi.org/10.13039/100000002</institution-id><institution>National Institutes of Health</institution></institution-wrap></funding-source><award-id>T32GM140935</award-id><principal-award-recipient><name><surname>Arp</surname><given-names>Nicholas L</given-names></name></principal-award-recipient></award-group><funding-statement>The funders had no role in study design, data collection and interpretation, or the decision to submit the work for publication.</funding-statement></funding-group><custom-meta-group><custom-meta specific-use="meta-only"><meta-name>Author impact statement</meta-name><meta-value>Chromosomal instability of cancer can be quantitatively measured by phylogenetic analysis of 200 tumor cells while using evolutionary principles to account for cellular selection.</meta-value></custom-meta></custom-meta-group></article-meta></front><body><sec id="s1" sec-type="intro"><title>Introduction</title><p>Chromosomal instability (CIN) is characterized by persistent whole-chromosome gain and loss through mis-segregation during cell division. Genome instability is a hallmark of cancer (<xref ref-type="bibr" rid="bib33">Hanahan and Weinberg, 2011</xref>) and one type, CIN, is the principal driver of aneuploidy, a feature of ~80% of solid tumors (<xref ref-type="bibr" rid="bib34">Hancock et al., 2004</xref>; <xref ref-type="bibr" rid="bib42">Knouse et al., 2017</xref>; <xref ref-type="bibr" rid="bib77">Weaver and Cleveland, 2006</xref>). CIN potentiates tumorigenesis (<xref ref-type="bibr" rid="bib25">Foijer et al., 2017</xref>; <xref ref-type="bibr" rid="bib46">Levine et al., 2017</xref>; <xref ref-type="bibr" rid="bib68">Silk et al., 2013</xref>) and associates with therapeutic resistance (<xref ref-type="bibr" rid="bib36">Ippolito et al., 2020</xref>; <xref ref-type="bibr" rid="bib44">Lee et al., 2011</xref>; <xref ref-type="bibr" rid="bib49">Lukow et al., 2020</xref>; <xref ref-type="bibr" rid="bib58">Pavelka et al., 2010</xref>), metastasis (<xref ref-type="bibr" rid="bib5">Bakhoum et al., 2018</xref>) and poor survival outcomes (<xref ref-type="bibr" rid="bib3">Bakhoum et al., 2011</xref>; <xref ref-type="bibr" rid="bib18">Denu et al., 2016</xref>; <xref ref-type="bibr" rid="bib37">Jamal-Hanjani et al., 2017</xref>). Thus, CIN is an important characteristic of cancer biology. Despite its importance, CIN has not emerged as a clinical biomarker, in part because it is challenging to quantify.</p><p>Although CIN has classically been characterized as binary—tumors either have it or not—recent evidence highlights the importance of the rate of chromosome mis-segregation and the specific aneuploidies it produces. For example, clinical outcomes partially depend on aneuploidy of specific chromosomes (<xref ref-type="bibr" rid="bib15">Davoli et al., 2013</xref>; <xref ref-type="bibr" rid="bib67">Sheltzer et al., 2017</xref>; <xref ref-type="bibr" rid="bib75">Vasudevan et al., 2020</xref>). Further, higher levels of CIN suppress tumor growth when they surpass a critical threshold, thought to be due to lethal loss of essential genes and irregular expression due to imbalanced gene dosage (<xref ref-type="bibr" rid="bib26">Funk et al., 2021</xref>; <xref ref-type="bibr" rid="bib68">Silk et al., 2013</xref>; <xref ref-type="bibr" rid="bib79">Weaver and Cleveland, 2008</xref>; <xref ref-type="bibr" rid="bib83">Zasadil et al., 2014</xref>). Moreover, baseline CIN may predict chemotherapeutic response to paclitaxel (<xref ref-type="bibr" rid="bib38">Janssen et al., 2009</xref>; <xref ref-type="bibr" rid="bib73">Swanton et al., 2009</xref>) and is proposed to both promote detection by or evasion from the immune system (<xref ref-type="bibr" rid="bib16">Davoli et al., 2017</xref>; <xref ref-type="bibr" rid="bib62">Santaguida et al., 2017</xref>). No single or standardized analytically valid measure of CIN has emerged and this gap has precluded its clinical validation as a prognostic or predictive biomarker.</p><p>Prior measures of CIN use various means to compare levels in tumors or populations, but do not establish a standardized quantitative rate. These prior measures include histologic analysis of mitotic defects (<xref ref-type="bibr" rid="bib3">Bakhoum et al., 2011</xref>; <xref ref-type="bibr" rid="bib39">Jin et al., 2020</xref>), fluorescence in situ hybridization (FISH) with probes to detect individual chromosomes (<xref ref-type="bibr" rid="bib74">Thompson and Compton, 2008</xref>), and gene-expression methodologies like CIN scores (<xref ref-type="bibr" rid="bib10">Carter et al., 2006</xref>). While these methods are readily accessible, they have significant drawbacks for clinical application. FISH and mitotic visualization approaches are laborious. Direct visualization of mitotic defects to measure CIN is only possible in the most proliferative tumors where enough cells are captured in short-lived mitosis. FISH typically quantifies only a subset of chromosomes, which will be misleading if there is bias toward specific chromosome gains/losses (<xref ref-type="bibr" rid="bib20">Dumont et al., 2020</xref>). While gene expression scores are proposed as indirect measures of CIN, they are not specific to CIN and correlate highly with proliferation and structural aneuploidy (<xref ref-type="bibr" rid="bib10">Carter et al., 2006</xref>; <xref ref-type="bibr" rid="bib66">Sheltzer, 2013</xref>).</p><p>Single-cell sequencing promises major advances in quantitative measures of CIN by displaying cell-cell variation for each chromosome across hundreds of cells (<xref ref-type="bibr" rid="bib52">Navin et al., 2011</xref>; <xref ref-type="bibr" rid="bib76">Wang et al., 2014</xref>). However, selection poses another complication. To date, single-cell analyses have identified surprisingly low cell-cell karyotype variation, even when mitotic errors are directly observed by microscopy (<xref ref-type="bibr" rid="bib8">Bolhaqueiro et al., 2019</xref>; <xref ref-type="bibr" rid="bib27">Gao et al., 2016</xref>; <xref ref-type="bibr" rid="bib41">Kim et al., 2018</xref>; <xref ref-type="bibr" rid="bib54">Nelson et al., 2020</xref>; <xref ref-type="bibr" rid="bib76">Wang et al., 2014</xref>). These observations highlight the confounding role of selection against aneuploid karyotypes in measuring CIN in human tumors. Indeed, selection reduces karyotype variance in cancer cell populations that directly exhibit mitotic errors (<xref ref-type="bibr" rid="bib29">Gerstung et al., 2020</xref>; <xref ref-type="bibr" rid="bib36">Ippolito et al., 2020</xref>; <xref ref-type="bibr" rid="bib49">Lukow et al., 2020</xref>). Here, we seek to overcome gap by modeling chromosomal instability and explicitly considering the evolutionary selection of aneuploid cells, to derive a quantitative measure.</p><p>We describe a quantitative framework to measure CIN by sampling population structure and cell-cell karyotypic variance in human tumors, accounting for selection on aneuploid karyotypes. We built our framework on the use of phylogenetic topology measures to quantify underlying evolutionary processes (<xref ref-type="bibr" rid="bib51">Mooers and Heard, 1997</xref>); in this case to quantify CIN from both the diversity and the aneuploid phylogeny within a tumor. Using an agent-based model of CIN, we determine how distinct types and degrees of selective pressure shape the karyotype distribution and population structure of tumor cells at different rates of chromosome mis-segregation. We then use this in silico model as a foundation for parameter inference to provide a quantitative estimate of CIN as the numerical rate of chromosome mis-segregation per cell division. We apply this model to quantify CIN caused by the chemotherapeutic paclitaxel in culture. Next, using existing single-cell whole-genome sequencing data (scDNAseq), we measure CIN in cancer biopsy and organoid samples. As a whole, this work provides a framework to quantify CIN in human tumors, a first step toward developing CIN as a prognostic and predictive biomarker.</p></sec><sec id="s2" sec-type="results"><title>Results</title><sec id="s2-1"><title>A framework for modeling CIN and karyotype selection</title><p>To assess intratumoral CIN via cell-cell karyotype heterogeneity, we considered how selection on aneuploid karyotypes impacts observed chromosomal heterogeneity within a tumor. By modeling fitness of aneuploid cells, we observe chromosomal variation in a population of surviving cells. The selective pressure of diverse and specific aneuploidies on human cells has not been, to our knowledge, directly measured. Therefore, we employ previously developed models of selection.</p><p>In models of CIN, fit karyotypes are selected while unfit aneuploid karyotypes are eliminated over time (<xref ref-type="bibr" rid="bib36">Ippolito et al., 2020</xref>; <xref ref-type="bibr" rid="bib60">Ravichandran et al., 2018</xref>; <xref ref-type="bibr" rid="bib67">Sheltzer et al., 2017</xref>; <xref ref-type="bibr" rid="bib75">Vasudevan et al., 2020</xref>). We use two previously proposed models of aneuploidy-associated cellular fitness, as well as hybrid and neutral selection models. The Gene Abundance model is based on the relatively low incidence of aneuploidy in normal tissues and assumes cellular fitness declines as the cell’s karyotype diverges from a balanced euploid karyotype (<xref ref-type="bibr" rid="bib65">Sheltzer and Amon, 2011</xref>; <xref ref-type="bibr" rid="bib84">Zhu et al., 2012</xref>). When an individual chromosome diverges from euploid balance (2 N, 3 N, 4 N, for example), its contribution to cellular fitness is weighted by its abundance of genes (<xref ref-type="fig" rid="fig1s1">Figure 1—figure supplement 1A</xref>, left). Alternatively, the Driver Density model assumes that each chromosome’s contribution to cellular fitness is weighted by its ratio of Tumor suppressor genes, Oncogenes, and Essential genes (TOEs)(<xref ref-type="bibr" rid="bib15">Davoli et al., 2013</xref>; <xref ref-type="bibr" rid="bib43">Laughney et al., 2015</xref>). For example, Driver Density selection will favor loss of chromosomes with many tumor suppressors and favor gain of chromosomes replete with oncogenes and essential genes (<xref ref-type="fig" rid="fig1s1">Figure 1—figure supplement 1A</xref>, right). The hybrid averaged model accounts for both karyotypic balance and TOE densities (<xref ref-type="fig" rid="fig1s1">Figure 1—figure supplement 1A</xref>, middle). Using these fitness models, we assigned chromosome scores to reflect each chromosome’s value to cellular fitness (<xref ref-type="fig" rid="fig1s1">Figure 1—figure supplement 1B</xref>, <xref ref-type="table" rid="table1">Table 1</xref>), the sum of which represent the total fitness value for the cell, relative to a value of 1 for a euploid cell. Further, we scaled the impact of cell fitness with a scaling factor, S, ranging from 0 (no selection) to 100 (high selection). While these models are approximations, they are nevertheless useful to estimate how mis-segregation and selective pressure cooperate to mold karyotypes in the cell population.</p><table-wrap id="table1" position="float"><label>Table 1.</label><caption><title>Base chromosome-specific fitness scores for individual models.</title></caption><table frame="hsides" rules="groups"><thead><tr><th align="left" valign="top"/><th align="left" valign="top">Selection model</th><th align="left" valign="top"/><th align="left" valign="top"/></tr></thead><tbody><tr><td align="left" valign="top">CHR ARM</td><td align="left" valign="top">Gene Abundance</td><td align="left" valign="top">Driver Density</td><td align="left" valign="top">Hybrid</td></tr><tr><td align="left" valign="top">1p</td><td align="left" valign="top">0.04780162</td><td align="left" valign="top">–0.0024018</td><td align="left" valign="top">0.02269992</td></tr><tr><td align="left" valign="top">1q</td><td align="left" valign="top">0.04340321</td><td align="left" valign="top">0.03244362</td><td align="left" valign="top">0.03792341</td></tr><tr><td align="left" valign="top">2p</td><td align="left" valign="top">0.02733655</td><td align="left" valign="top">0.02935717</td><td align="left" valign="top">0.02834686</td></tr><tr><td align="left" valign="top">2q</td><td align="left" valign="top">0.04244054</td><td align="left" valign="top">0.03943267</td><td align="left" valign="top">0.0409366</td></tr><tr><td align="left" valign="top">3p</td><td align="left" valign="top">0.02310412</td><td align="left" valign="top">0.03289695</td><td align="left" valign="top">0.02800053</td></tr><tr><td align="left" valign="top">3q</td><td align="left" valign="top">0.0299756</td><td align="left" valign="top">0.05416736</td><td align="left" valign="top">0.04207148</td></tr><tr><td align="left" valign="top">4p</td><td align="left" valign="top">0.01238195</td><td align="left" valign="top">0.01784909</td><td align="left" valign="top">0.01511552</td></tr><tr><td align="left" valign="top">4q</td><td align="left" valign="top">0.03181796</td><td align="left" valign="top">0.02901324</td><td align="left" valign="top">0.0304156</td></tr><tr><td align="left" valign="top">5p</td><td align="left" valign="top">0.01178443</td><td align="left" valign="top">0.04281166</td><td align="left" valign="top">0.02729805</td></tr><tr><td align="left" valign="top">5q</td><td align="left" valign="top">0.03787615</td><td align="left" valign="top">0.01949934</td><td align="left" valign="top">0.02868775</td></tr><tr><td align="left" valign="top">6p</td><td align="left" valign="top">0.02557719</td><td align="left" valign="top">0.02398619</td><td align="left" valign="top">0.02478169</td></tr><tr><td align="left" valign="top">6q</td><td align="left" valign="top">0.02554399</td><td align="left" valign="top">0.00011625</td><td align="left" valign="top">0.01283012</td></tr><tr><td align="left" valign="top">7p</td><td align="left" valign="top">0.0179588</td><td align="left" valign="top">0.09889284</td><td align="left" valign="top">0.05842582</td></tr><tr><td align="left" valign="top">7q</td><td align="left" valign="top">0.03231589</td><td align="left" valign="top">0.06933314</td><td align="left" valign="top">0.05082451</td></tr><tr><td align="left" valign="top">8p</td><td align="left" valign="top">0.01591728</td><td align="left" valign="top">0.02769564</td><td align="left" valign="top">0.02180646</td></tr><tr><td align="left" valign="top">8q</td><td align="left" valign="top">0.0254942</td><td align="left" valign="top">0.05861427</td><td align="left" valign="top">0.04205423</td></tr><tr><td align="left" valign="top">9p</td><td align="left" valign="top">0.01301266</td><td align="left" valign="top">–0.0012941</td><td align="left" valign="top">0.00585929</td></tr><tr><td align="left" valign="top">9q</td><td align="left" valign="top">0.02572657</td><td align="left" valign="top">0.04702681</td><td align="left" valign="top">0.03637669</td></tr><tr><td align="left" valign="top">10 p</td><td align="left" valign="top">0.0112201</td><td align="left" valign="top">–0.0364218</td><td align="left" valign="top">–0.0126008</td></tr><tr><td align="left" valign="top">10q</td><td align="left" valign="top">0.02750253</td><td align="left" valign="top">0.01142688</td><td align="left" valign="top">0.01946471</td></tr><tr><td align="left" valign="top">11 p</td><td align="left" valign="top">0.01961858</td><td align="left" valign="top">0.03818621</td><td align="left" valign="top">0.0289024</td></tr><tr><td align="left" valign="top">11q</td><td align="left" valign="top">0.03629936</td><td align="left" valign="top">0.01898784</td><td align="left" valign="top">0.0276436</td></tr><tr><td align="left" valign="top">12 p</td><td align="left" valign="top">0.0142575</td><td align="left" valign="top">0.0551551</td><td align="left" valign="top">0.0347063</td></tr><tr><td align="left" valign="top">12q</td><td align="left" valign="top">0.03659812</td><td align="left" valign="top">0.06273786</td><td align="left" valign="top">0.04966799</td></tr><tr><td align="left" valign="top">13 p</td><td align="left" valign="top">0</td><td align="left" valign="top">0</td><td align="left" valign="top">0</td></tr><tr><td align="left" valign="top">13q</td><td align="left" valign="top">0.02333649</td><td align="left" valign="top">–0.0101539</td><td align="left" valign="top">0.00659128</td></tr><tr><td align="left" valign="top">14 p</td><td align="left" valign="top">1.66E-05</td><td align="left" valign="top">0</td><td align="left" valign="top">8.30E-06</td></tr><tr><td align="left" valign="top">14q</td><td align="left" valign="top">0.03792594</td><td align="left" valign="top">0.02557439</td><td align="left" valign="top">0.03175016</td></tr><tr><td align="left" valign="top">15 p</td><td align="left" valign="top">0</td><td align="left" valign="top">0</td><td align="left" valign="top">0</td></tr><tr><td align="left" valign="top">15q</td><td align="left" valign="top">0.03701306</td><td align="left" valign="top">0.0206566</td><td align="left" valign="top">0.02883483</td></tr><tr><td align="left" valign="top">16 p</td><td align="left" valign="top">0.02383442</td><td align="left" valign="top">0.04334736</td><td align="left" valign="top">0.03359089</td></tr><tr><td align="left" valign="top">16q</td><td align="left" valign="top">0.01900446</td><td align="left" valign="top">–0.0071444</td><td align="left" valign="top">0.00593005</td></tr><tr><td align="left" valign="top">17 p</td><td align="left" valign="top">0.01548573</td><td align="left" valign="top">–0.0085975</td><td align="left" valign="top">0.00344414</td></tr><tr><td align="left" valign="top">17q</td><td align="left" valign="top">0.03553586</td><td align="left" valign="top">0.04363474</td><td align="left" valign="top">0.0395853</td></tr><tr><td align="left" valign="top">18 p</td><td align="left" valign="top">0.00627396</td><td align="left" valign="top">0.00533697</td><td align="left" valign="top">0.00580547</td></tr><tr><td align="left" valign="top">18q</td><td align="left" valign="top">0.01434049</td><td align="left" valign="top">–0.0263632</td><td align="left" valign="top">–0.0060113</td></tr><tr><td align="left" valign="top">19 p</td><td align="left" valign="top">0.02159372</td><td align="left" valign="top">0.05371416</td><td align="left" valign="top">0.03765394</td></tr><tr><td align="left" valign="top">19q</td><td align="left" valign="top">0.02813325</td><td align="left" valign="top">0.00550338</td><td align="left" valign="top">0.01681831</td></tr><tr><td align="left" valign="top">20 p</td><td align="left" valign="top">0.0089628</td><td align="left" valign="top">0.04351025</td><td align="left" valign="top">0.02623653</td></tr><tr><td align="left" valign="top">20q</td><td align="left" valign="top">0.01526996</td><td align="left" valign="top">0.04993593</td><td align="left" valign="top">0.03260295</td></tr><tr><td align="left" valign="top">21 p</td><td align="left" valign="top">0.00232369</td><td align="left" valign="top">0</td><td align="left" valign="top">0.00116185</td></tr><tr><td align="left" valign="top">21q</td><td align="left" valign="top">0.01233215</td><td align="left" valign="top">–0.0033092</td><td align="left" valign="top">0.00451147</td></tr><tr><td align="left" valign="top">22 p</td><td align="left" valign="top">0.00013278</td><td align="left" valign="top">0</td><td align="left" valign="top">6.64E-05</td></tr><tr><td align="left" valign="top">22q</td><td align="left" valign="top">0.02297134</td><td align="left" valign="top">–0.0051581</td><td align="left" valign="top">0.0089066</td></tr><tr><td align="left" valign="top">Xp</td><td align="left" valign="top">0.01555213</td><td align="left" valign="top">0</td><td align="left" valign="top">0.00777606</td></tr><tr><td align="left" valign="top">Xp</td><td align="left" valign="top">0.02499627</td><td align="left" valign="top">0</td><td align="left" valign="top">0.01249813</td></tr></tbody></table></table-wrap><p>We employed these selection models in an agent-based model of exponential population growth wherein each cell has its own karyotype (<xref ref-type="fig" rid="fig1">Figure 1</xref> and <xref ref-type="fig" rid="fig1s1">Figure 1—figure supplement 1</xref>). Briefly, simulations started with 100 euploid cells and were run in discrete time steps with variable rates of selective pressure, S, and rates of chromosome mis-segregation (P<sub>misseg</sub>, see definitions in <xref ref-type="table" rid="table2">Table 2</xref>). The rate—or probability—of mis-segregation events, P<sub>misseg</sub>, is the measure of CIN. During each time step, cells have a P<sub>division</sub> ( = 0.5 for euploid) chance of dividing. Each dividing cell has a P<sub>misseg</sub> chance of improper segregation of each chromosome. Segmental chromosome breaks occur with a probability P<sub>break</sub>, set at 0 or 0.5. After division, fitness (F) of each daughter is assessed. Cells are removed from the population if any given chromosome has copy number 0 or >6. The P<sub>division</sub> value of the remaining viable cells is adjusted by the cell’s fitness under selection (F<sup>S</sup>). Due to computational limitations, pseudo-Moran or Wright-Fisher models are employed to limit the modeled cell population (<xref ref-type="fig" rid="fig1s1">Figure 1—figure supplement 1C, D</xref>). These limits did not significantly affect the measures extracted from these populations (<xref ref-type="fig" rid="fig1s2">Figure 1—figure supplement 2</xref>). Thus, these models simulate an evolving population of aneuploid cells under given rates of CIN, P<sub>misseg</sub>, and models and strength of selection.</p><table-wrap id="table2" position="float"><label>Table 2.</label><caption><title>Parameters varied during agent-based modeling.</title></caption><table frame="hsides" rules="groups"><thead><tr><th align="left" valign="top">Parameter</th><th align="left" valign="top">Description</th></tr></thead><tbody><tr><td align="left" valign="top">Pmisseg</td><td align="left" valign="top">Probability of mis-segregation per chromosome per division</td></tr><tr><td align="left" valign="top">Pbreak</td><td align="left" valign="top">Probability of chromosome breakage after mis-segregation</td></tr><tr><td align="left" valign="top">Pdivision</td><td align="left" valign="top">Probability of cellular division per time step</td></tr><tr><td align="left" valign="top">S</td><td align="left" valign="top">Magnitude of selective pressure on aneuploid karyotypes</td></tr></tbody></table></table-wrap><fig-group><fig id="fig1" position="float"><label>Figure 1.</label><caption><title>A framework for modeling CIN and karyotype selection.</title><p>(<bold>A</bold>) Chromosome arm scores for each model of karyotype selection. Gene Abundance scores are derived from the number of genes per chromosome arm normalized to the number of all genes. Chromosome arms 13 p and 15 p did not have an abundance score and were set to 0. Driver Density scores come from the pan-cancer chromosome arm scores derived in <xref ref-type="bibr" rid="bib15">Davoli et al., 2013</xref>, and normalized to the sum of chromosome arm scores for chromosomes 1-22,X. Chromosome arms 13 p, 14 p, 15 p, 21 p, 22 p, and chromosome X did not have driver scores and were set to 0. Hybrid model scores are set to the average of the Driver and Abundance models. The neutral model (not displayed) is performed with all cell’s fitness constitutively equal to 1 regardless of karyotype. (<bold>B</bold>) Framework for the simulation of and selection on cellular populations with CIN. Cells divide (Pdivision starts at 0.5 in the exponential pseudo-Moran model and is constitutively equal to 1 for the constant Wright-Fisher model) and probabilistically mis-segregate chromosomes (Pmisseg ∈ [0, 0.001… 0.05]). After, cells experience selection under one of the selection models, altering cellular fitness and the probability (Pdivision) a cell will divide again (green check). Additionally, cells wherein the copy number of any chromosome falls to zero or surpasses 6 are removed (red x). After this, the cycle repeats. See Materials and methods for further details.</p></caption><graphic mime-subtype="jpeg" mimetype="image" xlink:href="elife-69799.xml.media/fig1.jpg"/></fig><fig id="fig1s1" position="float" specific-use="child-fig"><label>Figure 1—figure supplement 1.</label><caption><title>Expanded model of chromosome mis-segregation and karyotypic selection.</title><p>Models of selection on aneuploid karyotypes. <bold>Left</bold>. In the Gene Abundance model, chromosomes that encode a larger number of genes contribute more to cellular fitness (F). Thus, large chromosomes have a higher fitness score (f<sub>c</sub>). Deviation from the average ploidy of the population results in a reduced Contextual Fitness Score (CFS) for each chromosome, the sum of which represents the fitness of the cell. <bold>Right</bold>. In the Driver Density Model, the fitness contribution of a chromosome depends on the ratio of oncogenes and essential genes to tumor suppressors (OG-ESG:TSG). Gaining chromosomes with a higher OG-ESG:TSG ratio provides a fitness advantage while gaining more suppressive chromosomes invokes a fitness cost. These scores are still normalized to the ploidy of the average ploidy of the population to ensure that higher ploidy populations are not arbitrarily more fit. <bold>Middle</bold>. The Hybrid model takes the average of the fitness scores calculated in the other models. The neutral selection model (not shown) treats all karyotypes as equally fit. Base chromosome arm fitness scores for each model. Only the Hybrid and Driver Density model have negatively scored chromosomes, meaning their loss provides a fitness benefit. The neutral selection model does not require chromosome arm fitness scores. Simulating CIN in exponentially growing populations with pseudo-Moran limits. (<bold>0</bold>) Populations are founded by 100 founder cells and the simulation is initiated. (<bold>1</bold>) CFS values are calculated for each chromosome in a cell according to the chosen model. (<bold>2</bold>) Cellular fitness is calculated based on CFS values. (<bold>3</bold>) Selective pressure (S) is applied on cellular fitness values (F). (<bold>4</bold>) Cells are checked to see if any death conditions are met and if the population limit is met. (<bold>5</bold>) Cells probabilistically enter mitosis if their fitness value exceeds a random float (R) between 0 and 2. Thus P<sub>division</sub> = P(F<sub>M</sub> >R). If a cell does not divide, it skips the next step. (<bold>6</bold>) If a cell enters mitosis, each chromosome has an opportunity to mis-segregate probabilistically. For each chromosome, a mis-segregation occurs if a random float (R), from 0 to 1, falls below P<sub>misseg</sub>. After a chromosome mis-segregation is determined, the chromosome arms may be individually segregated (i.e. reciprocal CNA) if a random float (R), from 0 to 1, falls below P<sub>break</sub>. The cycle repeats and new CFS values are calculated, unless (<bold>7</bold>) stop conditions are met. When populations reach or exceed 3500 cells, a random half of the population is eliminated and the remaining cells continue the cycle. Simulating CIN in constant-size populations with Wright-Fisher dynamics. (<bold>0</bold>) Populations are initiated by 4500 euploid cells which (<bold>1</bold>) divide every step. (<bold>2</bold>) Chromosomes are mis-segregated as in the exponential pseudo-Moran model described above. (<bold>3</bold>) If stop conditions are met, the simulation ends and data are exported. If the cycle continues, (<bold>4</bold>) CFS values are calculated and used to (<bold>5</bold>) determine cellular fitness, after which, (<bold>6</bold>) selective pressure is applied. (<bold>7</bold>) Cells die if they lose both copies of a chromosome or exceed the upper limit of six. Additionally, to approximate Wright-Fisher dynamics, cells die if 1/(F<sup>S</sup> +0.001) exceeds a random float from 0 to 5. Thus, the baseline rate of cell death is ~0.2. (<bold>8</bold>) Each chromosome copy number is stored and the population is re-initiated with 4500 new cells. The copy numbers for each of new cell’s chromosomes are randomly and independently drawn from the copy number distributions of the previous generation. The cycle then repeats until the simulation ends (step 3).</p></caption><graphic mime-subtype="jpeg" mimetype="image" xlink:href="elife-69799.xml.media/fig1-figsupp1.jpg"/></fig><fig id="fig1s2" position="float" specific-use="child-fig"><label>Figure 1—figure supplement 2.</label><caption><title>Population growth limits do not bias population measures.</title><p>(<bold>A</bold>) Growth curves of populations simulated under the Hybrid selection model and exponential pseudo-Moran growth model with S ∈[0,1] and Pmisseg <sub>misseg</sub> = 0.022 and limited to 3000, 6000, and 24,000 cells (n = 4 simulations each). (<bold>B</bold>) MKV (normalized to mean ploidy of the population) values steadily increase over time. (<bold>C</bold>) Loess regression curves show no significant deviations based on the population threshold, regardless of selection. Tree-tip-normalized Sackin index values for each population over time. No significant deviations based on the population threshold, regardless of selection.</p></caption><graphic mime-subtype="jpeg" mimetype="image" xlink:href="elife-69799.xml.media/fig1-figsupp2.jpg"/></fig></fig-group></sec><sec id="s2-2"><title>Evolutionary dynamics is imparted by CIN</title><p>To understand the interplay between CIN and selection, we simulated 100 steps of cell growth with CIN under each selection model. We varied the rate of CIN (P<sub>misseg,c</sub> ∈[0, 0.001… 0.05] per chromosome; or 0–2.3 chromosome mis-segregations per division) and selective pressure ranging from none to heavy selection (S ∈[0, 2… 100]). As expected, the simulated cell number increases rapidly to the pseudo-Moran cap of 3000, where it remains (<xref ref-type="fig" rid="fig2">Figure 2A</xref>). As displayed in <xref ref-type="fig" rid="fig2">Figure 2B</xref>, diversity of the cell population, expressed as mean karyotypic variance increases over time, but also depends on mis-segregation rate, and selection levels (<xref ref-type="fig" rid="fig2">Figure 2B</xref>). As expected, high mis-segregation rates (P<sub>misseg</sub>, Y axis) and low selection (S = 0; top row) enhance the variance of the population. Further, without selection (S = 0; top row) all models returned comparable profiles over time, resembling neutral selection. However, when selective pressure is applied (S > 0), the distinct profiles appear. The abundance model (first column) negatively selects against all aneuploid karyotypes and yields low heterogeneity that increases modestly with mis-segregation rate. With the Driver model (second column), there is a sharp increase in heterogeneity even at low mis-segregation rates, as this model favors specific aneuploid states that maximizes oncogenes and minimizes tumor suppressors. The Hybrid model falls between the other two. Results were not specific to the pseudo-Moran process of capping at 3000 cells—dynamics were similar in the constant-population Wright-Fisher model (<xref ref-type="fig" rid="fig2s1">Figure 2—figure supplement 1A, B</xref>). These data illustrate how CIN and selection operate together to shape the karyotype diversity in the cell population.</p><fig-group><fig id="fig2" position="float"><label>Figure 2.</label><caption><title>Evolutionary dynamics imparted by CIN.</title><p>(<bold>A</bold>) Population growth curve in the absence of selective pressure (P<sub>misseg</sub> = 0.001, S = 0, n = 3 simulations). The steady state population in null selection conditions is 3000 cells. (<bold>B</bold>) Heatmaps depicting dynamics of karyotype diversity as a function of time (steps), mis-segregation rate (P<sub>misseg</sub>), and selection (S) under each model of selection. Columns represent the same model; rows represent the same selection level. Mean karyotype diversity (MKV) is measured as the variance of each chromosome averaged across all chromosomes 1–22, and chromosome X. Low and high MKV are shown in white and blue respectively (n = 3 simulations for every combination of parameters). (<bold>C</bold>) Population growth under each model, varying P<sub>misseg</sub> and S. P<sub>misseg</sub>∈ [0.001, 0.022, 0.050] translate to about 0.046, 1, and 2.3 mis-segregations per division respectively for diploid cells. (<bold>D</bold>) Dynamics of the average ploidy (total # chromosome arms / 46) of a population while varying P<sub>misseg</sub> and S. (<bold>E</bold>) Dynamics of ploidy under each model for diploid and tetraploid founding populations. P<sub>misseg</sub>∈ [0.01, 0.02] translate to about 0.46 and 0.92 mis-segregations for diploid cells and 0.92 and 1.84 mis-segregations for tetraploid cells. (<bold>F</bold>) Fitness (F<sup>S</sup>) over time for diploid and tetraploid founding populations evolved under each model. (<bold>G</bold>) Karyotype diversity dynamics for diploid and tetraploid founding populations. MKV is normalized to the mean ploidy of the population at each time step. Plotted lines in C-G are local regressions of n = 3 simulations.</p></caption><graphic mime-subtype="jpeg" mimetype="image" xlink:href="elife-69799.xml.media/fig2.jpg"/></fig><fig id="fig2s1" position="float" specific-use="child-fig"><label>Figure 2—figure supplement 1.</label><caption><title>Chromosomal instability and karyotype selection in constant-size populations approximating Wright-Fisher dynamics.</title><p>(<bold>A</bold>) Population size over time in the absence of selective pressure (P<sub>misseg</sub> = 0.001, S = 0, n = 3 simulations). The steady state population in null selection conditions is ~3600 cells as data is exported before populations are re-initiated. Dashed line represents the population at (re-)initiation (4500 cells). (<bold>B</bold>) Heatmaps depicting dynamics of karyotype diversity as a function of time (steps), mis-segregation rate (P<sub>misseg</sub>), and selection (S) under each model of selection. Columns represent the same model; rows represent the same selection level. Mean karyotype diversity (MKV) is measured as the variance of each chromosome averaged across all chromosomes 1–22, and chromosome X. Low and high MKV are shown in white and green respectively (n = 3 simulations for every combination of parameters). (<bold>C</bold>) Population growth under each model, varying P<sub>misseg</sub> and S. P<sub>misseg</sub>∈ [0.001, 0.022, 0.050] translate to about 0.046, 1, and 2.3 mis-segregations per division respectively for diploid cells. Top dashed line represents the population at (re-)initiation (4500 cells). Bottom dashed line represents the steady state population in selection-null conditions. (<bold>D</bold>) Dynamics of the average ploidy (total # chromosome arms / 46) of a population while varying P<sub>misseg</sub> and S. (<bold>E</bold>) Dynamics of ploidy under each model for diploid and tetraploid founding populations. P<sub>misseg</sub>∈ [0.01, 0.02] translate to about 0.46 and 0.92 mis-segregations for diploid cells and 0.92 and 1.84 mis-segregations for tetraploid cells. (<bold>F</bold>) Fitness (F<sup>S</sup>) over time for diploid and tetraploid founding populations evolved under each model. (<bold>G</bold>) Karyotype diversity dynamics for diploid and tetraploid founding populations. MKV is normalized to the mean ploidy of the population at each time step. Plotted lines in C-G are local regressions of n = 3 simulations.</p></caption><graphic mime-subtype="jpeg" mimetype="image" xlink:href="elife-69799.xml.media/fig2-figsupp1.jpg"/></fig><fig id="fig2s2" position="float" specific-use="child-fig"><label>Figure 2—figure supplement 2.</label><caption><title>Fitness of diploid and tetraploid CIN +populations.</title><p>(<bold>A</bold>) Fitness landscape of simulations founded by diploid cells under exponential pseudo-Moran growth dynamics. (<bold>B</bold>) Size of simulated populations founded by diploid cells under exponential pseudo-Moran growth dynamics. (<bold>C</bold>) Fitness landscape of simulations founded by diploid cells under constant Wright-Fisher growth dynamics. (<bold>D</bold>) Size of simulated populations founded by diploid cells under constant Wright-Fisher growth dynamics. (<bold>E</bold>) Fitness landscape of simulations founded by tetraploid cells under exponential pseudo-Moran growth dynamics. (<bold>F</bold>) Size of simulated populations founded by tetraploid cells under exponential pseudo-Moran growth dynamics. (<bold>G</bold>) Fitness landscape of simulations founded by tetraploid cells under constant Wright-Fisher growth dynamics. (<bold>H</bold>) Size of simulated populations founded by tetraploid cells under constant Wright-Fisher growth dynamics.</p></caption><graphic mime-subtype="jpeg" mimetype="image" xlink:href="elife-69799.xml.media/fig2-figsupp2.jpg"/></fig></fig-group><p>High levels of selection against aneuploid cells are expected to impede cell growth. To visualize this, we quantified the population of viable cells with distinct models (<xref ref-type="fig" rid="fig2">Figure 2C</xref>). As expected with the Abundance model at S = 10 and S = 100, cells proliferated more slowly with higher rates of mis-segregation. By contrast, the Driver model saw no growth defect as they favored specific aneuploid states that are easily reached with missegregation. As before, the Hybrid model, is intermediate, and findings are not impacted by pseudo-Moran or Wright-Fisher restrictions on cell number (<xref ref-type="fig" rid="fig2s1">Figure 2—figure supplement 1C</xref>).</p><p>To further assess model dynamics, we examined time-course of average cellular ploidy—the number of chromosomes divided by 23. In many cases, the mean ploidy of the populations tend to increase over time (<xref ref-type="fig" rid="fig2">Figure 2D</xref>, <xref ref-type="fig" rid="fig2s1">Figure 2—figure supplement 1D</xref>), particularly in the absence of selection (S = 0; top). This is likely due to a higher permissiveness to chromosome gains than losses in our model (since cells ‘die’ with nullisomy or any chromosome >6, the optimum is 3.0). With selection (S = 10; S = 100 rows), the models diverge. In the abundance model, populations remain near diploid. With the Driver model, the average ploidy increases more rapidly due to favoring aneuploidy states that favor high oncogenes and low tumor suppressors, consistent with previous computational models built on chromosome-specific driver densities (<xref ref-type="bibr" rid="bib15">Davoli et al., 2013</xref>; <xref ref-type="bibr" rid="bib43">Laughney et al., 2015</xref>). Under the Hybrid model, ploidy increases modestly. Similar effects are seen with the constant-population Wright-Fisher model (<xref ref-type="fig" rid="fig2s1">Figure 2—figure supplement 1D</xref>). In sum, selection and mis-segregation cooperate to shape the aneuploid karyotypes diversity, cell proliferation and average ploidy in a population of cells, or a human tumor. Further, sampling karyotypes in a cell population does not allow direct determination of mis-segregation rates, as their diversity is influenced by other factors such as selective pressure, selection modality, and time.</p><p>In some tumors, genome doubling occurs early in tumor initiation relative to other copy number changes (<xref ref-type="bibr" rid="bib7">Bielski et al., 2018</xref>; <xref ref-type="bibr" rid="bib29">Gerstung et al., 2020</xref>). Genome doubling is accomplished, for example, by endoreduplication, by failed cytokinesis, or by cell-cell fusion. Genome doubling buffers against loss of chromosomes and thereby favors aneuploidy. To determine how genome doubling impacts evolution in our model, we compared diploid and tetraploid founders (<xref ref-type="fig" rid="fig2">Figure 2E–G</xref>). Both diploids and tetraploids tend to converge toward the near-triploid state (ploidy ~3), as observed in many human cancers (<xref ref-type="bibr" rid="bib11">Carter et al., 2012</xref>), although this is restrained to a degree with the Abundance and Hybrid models. Compared with diploid cells, tetraploidy buffered against the negative effects of cellular fitness in the Abundance model, despite generating similar levels of diversity over time (<xref ref-type="fig" rid="fig2">Figure 2F and G</xref>)— this is more pronounced when comparing P<sub>misseg</sub> = 0.1 in tetraploids versus P<sub>misseg</sub> = 0.2 in diploids to match the number of chromosome mis-segregations per division. This is consistent with the idea that tetraploidy serves as an intermediate enabling a near-triploid karyotype that is common in many cancers (<xref ref-type="bibr" rid="bib7">Bielski et al., 2018</xref>; <xref ref-type="bibr" rid="bib47">López et al., 2020</xref>). By contrast, in the Driver model, tetraploidy did not provide a selective advantage to high-CIN tumors (<xref ref-type="fig" rid="fig2">Figure 2F</xref>). Similar fitness, karyotype diversities, and ploidy increases were obtained with a Wright-Fisher model of population growth (<xref ref-type="fig" rid="fig2s1">Figure 2—figure supplement 1E-G</xref>, <xref ref-type="fig" rid="fig2s2">Figure 2—figure supplement 2</xref>).</p><p>Taken together, the agent-based model recapitulates expected key aspects of tumor evolution, lending credence to our model. Further, they illustrate the difficulty of inferring mis-segregation rates directly from assessing variation in karyotypes in human cancer. Nevertheless, this model provides a framework to incorporate selection to measure CIN through quantitative inference from the observed karyotypes, as we will demonstrate.</p></sec><sec id="s2-3"><title>Long-term karyotype diversity depends profoundly on selection modality</title><p>Some current measures of CIN are derived from karyotype diversity in the population. Yet, our model suggests that selection pressure will profoundly shape this diversity. To further understand the nature of karyotype diversity under selection, we evaluated their long-term dynamics, whether they exhibit clonality, and whether populations simulated under each model converge on a common karyotype.</p><p>We simulated diploid and tetraploid populations for 3000 time steps at a fixed mis-segregation rate, in an experimentally reported range, allowing for fragmentation of chromosome arms (P<sub>misseg</sub> = 0.003, P<sub>break</sub> = 0.5) (<xref ref-type="bibr" rid="bib2">Bakhoum et al., 2009</xref>; <xref ref-type="bibr" rid="bib8">Bolhaqueiro et al., 2019</xref>; <xref ref-type="bibr" rid="bib78">Weaver et al., 2007</xref>) and S ∈ [1,25] (<xref ref-type="fig" rid="fig3">Figure 3A</xref>). We visualized copy-number heatmaps indicating karyotypes of sampled cells from the population. As expected, population diversity is limited under the Abundance model (<xref ref-type="fig" rid="fig3">Figure 3B</xref>). Even after 3000 time steps, only a small number of unique alterations and sub-clonal alterations ( + 13 p/–15 p/–22 p) existed, likely passenger alterations as they offer no fitness advantage in this model. Moreover, the karyotype average of 1500 cells across five replicates resembled a diploid karyotype (<xref ref-type="fig" rid="fig3">Figure 3C</xref>, row 1), indicating that the Abundance model provides stabilizing selection around the euploid karyotype. In fact, populations simulated under this model with elevated selection (S = 25) quickly reach a low, steady-state level of karyotype diversity and fitness while those with the unmodified selection values (S = 1) take a longer time to reach this steady-state and have similar levels of karyotype diversity and fitness as the other models (<xref ref-type="fig" rid="fig3s1">Figure 3—figure supplement 1</xref>). To identify any contingencies that may affect these associations, we performed the same simulation using several variants of our model. We found this steady state to be consistent for tetraploid cells as well as when we eased the upper ploidy constraint from nc <sub>c</sub> = 6 to an extreme nc <sub>c</sub> = 10, when we imposed a severe, 90% fitness reduction for all cells with a haploidy, and when we simulated populations under the Wright-Fisher model (<xref ref-type="fig" rid="fig3">Figure 3C</xref>, rows 2–4).</p><fig-group><fig id="fig3" position="float"><label>Figure 3.</label><caption><title>Karyotype diversity depends profoundly on selection modality.</title><p>(<bold>A</bold>) Simulation scheme to assess long-term dynamics of karyotype evolution and karyotype convergence. (<bold>B</bold>) Heatmaps depicting the chromosome copy number profiles of a subset (n = 30 out of 300 sampled cells) of the simulated population with early CIN over time under each model of karyotype selection. (<bold>C</bold>) Average heatmaps (lower) show the average copy number across the 5 replicates for (1) the Exponential Psuedo-Moran (Base), (2) the base model with the upper copy number limit set to 10, (3) the base model that invokes a F<sub>M</sub> x 0.1 penalty for any cell with a haploid chromosome, (4) and the Constant Population-Size Wright-Fisher model. P<sub>misseg</sub> = 0.003; S = 25 (except Neutral model; S = 0); ploidy = 2.</p></caption><graphic mime-subtype="jpeg" mimetype="image" xlink:href="elife-69799.xml.media/fig3.jpg"/></fig><fig id="fig3s1" position="float" specific-use="child-fig"><label>Figure 3—figure supplement 1.</label><caption><title>Modeled population measures tracked over time.</title><p>(<bold>A</bold>) Average population ploidy over time for each selection model within each model variation. Data represent the mean and range (vertical lines) across five replicates for every 50 time steps in diploid populations with low selective pressure (light red) and high selective pressure (dark red) and tetraploid populations with low selective pressure (light blue) and high selective pressure (dark blue). (<bold>B</bold>) Average population fitness (log10) over time for each selection model within each model variation. Data represent the mean and range (vertical lines) across five replicates for every 50 time steps in diploid populations with low selective pressure (light red) and high selective pressure (dark red) and tetraploid populations with low selective pressure (light blue) and high selective pressure (dark blue). (<bold>C</bold>) Mean karyotype variance over time for each selection model within each model variation. Data represent the mean and range (vertical lines) across five replicates for every 50 time steps in diploid populations with low selective pressure (light red) and high selective pressure (dark red) and tetraploid populations with low selective pressure (light blue) and high selective pressure (dark blue).</p></caption><graphic mime-subtype="jpeg" mimetype="image" xlink:href="elife-69799.xml.media/fig3-figsupp1.jpg"/></fig></fig-group><p>The Driver Density and Hybrid models generate much more diversity (<xref ref-type="fig" rid="fig3">Figure 3B</xref>) but nevertheless converge by 3,000 timesteps (<xref ref-type="fig" rid="fig3s1">Figure 3—figure supplement 1</xref>). Without selection (neutral model), there is high diversity and no convergence over time. Taken together, these demonstrate a high dependence on the model of selection. However, the models are not highly dependent on ploidy constraints, haploid penalties, or on selection of Pseudo-Moran or Wright-Fisher restriction of cell numbers. Taken together, long-term populations are strongly shaped by the model of karyotype selection for a given P<sub>misseg</sub>, but relatively insensitive to other particular features of the model. This justifies our approach henceforth of varying only the selection model, the degree of selection (S), and P<sub>misseg</sub> to infer parameters from data via phylogenetic topology and Bayesian inference.</p></sec><sec id="s2-4"><title>Topological features of simulated phylogenies delineate CIN rate and karyotype selection</title><p>Given a model capable of recapitulating diversity and selective pressures, next we wish to infer P<sub>misseg</sub> as a measure of CIN from an observed population of cells. Phylogenetic trees provide insights into evolutionary processes of genetic diversification and selection. Moreover, the topology of the phylogenetic tree has been used as a quantitative measure of the underlying evolutionary processes (<xref ref-type="bibr" rid="bib12">Colijn and Plazzotta, 2018</xref>; <xref ref-type="bibr" rid="bib17">Dayarian and Shraiman, 2014</xref>; <xref ref-type="bibr" rid="bib50">Manceau et al., 2015</xref>; <xref ref-type="bibr" rid="bib53">Neher et al., 2014</xref>; <xref ref-type="bibr" rid="bib63">Scott et al., 2020</xref>).</p><p>Here, chromosome mis-segregation gives rise to karyotype heterogeneity, and the population of cells is then shaped by selection. To evaluate this, we use chromosome copy number-based phylogenetic reconstruction, since mutation rates are not high enough in tumors to reliably infer cellular relationships, particularly with low-copy sequencing. Once phylogenies are reconstructed from simulated and experimental populations, the topological features phylogenies can be compared. These features include ‘cherries’—two tips that share a direct ancestor—and ‘pitchforks—a clade with three tips (<xref ref-type="fig" rid="fig4">Figure 4A</xref>). Additionally, we considered a broader metric of topology, the Colless index, which measures the imbalance or asymmetry of the entire tree. To understand how these measures are affected by selection in simulated populations, we reconstructed phylogenies from 300 random cells from each population simulated with a range of selective pressures taken at 60 time steps (~30 divisions under Hybrid selection; <xref ref-type="fig" rid="fig4">Figure 4B</xref>). As seen previously, aneuploidy and mean karyotypic variance (MKV) decrease with selective pressure, a trend that is robust at high mis-segregation rates (<xref ref-type="fig" rid="fig4">Figure 4C</xref>). By contrast, Colless indices increase with mis-segregation rates and selective pressures, as the resulting variation and selection generate phylogenetic asymmetry. Accordingly, this imbalance is apparent in phylogenetic reconstructions of simulated populations (<xref ref-type="fig" rid="fig4">Figure 4D</xref>). Cherries, by contrast, decrease with selection due to selection against many aneuploidies (<xref ref-type="fig" rid="fig4">Figure 4C</xref>). Pitchforks seemed less informative. Therefore, we tentatively selected 4 phylogenetic parameters that can retain information about chromosome missegregation—aneuploidy, MKV, Colless, and Cherries.</p><fig id="fig4" position="float"><label>Figure 4.</label><caption><title>Topological features of simulated phylogenies delineate CIN rate and karyotype selection.</title><p>(<bold>A</bold>) Quantifiable features of karyotypically diverse populations. Heterogeneity between and within karyotypes is described by MKV and aneuploidy (inter- and intra-karyotype variance, see Materials and methods). We also quantify discrete topological features of phylogenetic trees, such as cherries (tip pairs) and pitchforks (3-tip groups), and a whole-tree measure of imbalance (or asymmetry), the Colless index. (<bold>B</bold>) Scheme to test how CIN and selection influence the phylogenetic topology of simulated populations. (<bold>C</bold>) Computed heterogeneity (aneuploidy and MKV) and topology (Colless index, cherries, pitchforks) summary statistics under varying P<sub>misseg</sub> and S values. MKV is normalized to the average ploidy of the population. Topological measures are normalized to population size. Spearman rank correlation coefficients (r) and p-values are displayed (n = 8 simulations). (<bold>D</bold>) Representative phylogenies for each hi/low CIN, hi/low selection parameter combination and their computed summary statistics. Each phylogeny represents n = 50 out of 300 cells for each simulation. (<bold>E</bold>) Dimensionality reduction of all simulations for each hi/low CIN, hi/low selection parameter combination using measures of karyotype heterogeneity only (left; MKV and aneuploidy) or measures of karyotype heterogeneity and phylogenetic topology (right; MKV, aneuploidy, Colless index, cherries, and pitchforks).</p></caption><graphic mime-subtype="jpeg" mimetype="image" xlink:href="elife-69799.xml.media/fig4.jpg"/></fig><p>To characterize how well the four measures retain information about the simulation parameters, we performed dimensionality reduction with measures of karyotype heterogeneity alone (MKV and aneuploidy) alone and adding Colless and cherries—measures of phylogenetic topology (<xref ref-type="fig" rid="fig4">Figure 4E</xref>). This analysis indicates that when considering heterogeneity alone simulations performed under high CIN/high selection (yellow) and low CIN/low selection (red) associate closely, meaning these measures of heterogeneity are not sufficient to distinguish these disparate conditions (<xref ref-type="fig" rid="fig4">Figure 4E</xref>, left). These similarities arise because high selection can mask the heterogeneity expected from high CIN. By contrast, combining measures of heterogeneity with those of phylogenetic topology can discriminate between simulations with disparate levels of CIN and selection (<xref ref-type="fig" rid="fig4">Figure 4E</xref>, right). This provides further evidence that measures of heterogeneity alone are not sufficient to infer CIN due to the confounding effects of selection, particularly when the nature of selection is unclear or can vary. Together these results indicate that phylogenetic topology preserves information about underlying levels of selective pressure and rates of chromosome mis-segregation. Further, phylogenetic topology of single-cell populations may be a suitable way to correct for selective pressure when estimating the rate of chromosome mis-segregation from measures of karyotype diversity.</p></sec><sec id="s2-5"><title>Experimental chromosome mis-segregation measured by Bayesian inference</title><p>To experimentally validate quantitative measures of CIN, we generated a high rate of chromosome mis-segregation with a clinically relevant concentration of paclitaxel (Taxol) over 48 hr (<xref ref-type="fig" rid="fig5">Figure 5A</xref>). We treated CAL51 breast cancer cells with either a DMSO control or 20 nM paclitaxel, which generates widespread aneuploidy due to chromosome mis-segregation on multipolar mitotic spindles (<xref ref-type="bibr" rid="bib83">Zasadil et al., 2014</xref>), verified in this experiment (<xref ref-type="fig" rid="fig5s1">Figure 5—figure supplement 1A</xref>). At 48 hr cells will have undergone 1–2 mitoses and, consistent with abnormal chromosome segregation, we observe broadened DNA content distributions by flow cytometry (<xref ref-type="fig" rid="fig5s1">Figure 5—figure supplement 1B</xref>). Using low-coverage scDNAseq data, we characterized the karyotypes of 36 DMSO- and 134 paclitaxel-treated cells. As expected, virtually all cells had extensive aneuploidy after paclitaxel, in contrast with low variance in the control (<xref ref-type="fig" rid="fig5">Figure 5B</xref>). Additionally, the mean of the resultant aneuploid karyotypes for each chromosome still resembled those of bulk-sequenced cells, highlighting that bulk-sequencing is an ensemble average, and does not detect variation in population aneuploidy, particularly with balanced mis-segregation events (<xref ref-type="fig" rid="fig5">Figure 5B</xref>, single-cell mean and bulk). In quantifying the absolute deviation from the modal control karyotype in each cell, and assuming a single mitosis, cells exposed to 20 nM paclitaxel mis-segregate 18.5 ± 0.5—a P<sub>misseg</sub> of ~0.42 considering the control’s sub-diploid modal karyotype (<xref ref-type="fig" rid="fig5">Figure 5C</xref>). The majority of these appeared to be whole-chromosome mis-segregations (<xref ref-type="fig" rid="fig5s2">Figure 5—figure supplement 2</xref>).</p><fig-group><fig id="fig5" position="float"><label>Figure 5.</label><caption><title>Experimental chromosome mis-segregation measured by Bayesian inference experimental scheme.</title><p>(<bold>A</bold>) Cal51 cells were treated with either DMSO or 20 nM paclitaxel for 48 hr prior to further analysis by time lapse imaging, bulk DNA sequencing, and scDNAseq. (<bold>B</bold>) Heatmaps showing copy number profiles derived from scDNAseq data, single-cell copy number averages, and bulk DNA sequencing. (<bold>C</bold>) Observed mis-segregations calculated as the absolute sum of deviations from the observed modal karyotype of the control. (<bold>D</bold>) Dimensionality reduction analysis of population summary statistics (aneuploidy, MKV, Colless index, cherries) from the first three time steps of all simulations performed under the Hybrid model. (<bold>E</bold>) 2D density plot showing joint posterior distributions from ABC analysis using population summary statistics computed from the paclitaxel-treated cells using the following priors and parameters: Growth Model = ‘exponential pseudo-Moran’, Selection Model = ‘Hybrid, initial ploidy = 2, 2 time steps, S ∈[0, 2… 100], P<sub>misseg</sub>∈[0, 0.005… 1.00] and a tolerance threshold of 0.05 to reject dissimilar simulation results. (see Materials and Methods). Vertical dashed line represents the experimentally observed mis-segregation rate. White + represents the mean of inferred values.</p></caption><graphic mime-subtype="jpeg" mimetype="image" xlink:href="elife-69799.xml.media/fig5.jpg"/></fig><fig id="fig5s1" position="float" specific-use="child-fig"><label>Figure 5—figure supplement 1.</label><caption><title>Induction of extensive chromosome mis-segregation via paclitaxel.</title><p>(<bold>A</bold>) Immunofluorescence time lapse montage of control Cal51 cells undergoing normal mitosis (top) and paclitaxel-treated treated cells undergoing a multipolar anaphase (middle) and partial cytokinesis failure (bottom). (<bold>B</bold>) Cell cycle profiles from flow cytometric analysis of Cal51 cells treated with either DMSO (72 hr) or 20 nM paclitaxel for 24, 48, or 72 hr. For FACS, cells treated for 48 hr were sorted into individual wells of 96-well plates. Sorting gate is shown by the red, dashed line.</p></caption><graphic mime-subtype="jpeg" mimetype="image" xlink:href="elife-69799.xml.media/fig5-figsupp1.jpg"/></fig><fig id="fig5s2" position="float" specific-use="child-fig"><label>Figure 5—figure supplement 2.</label><caption><title>Copy number profiles of DMSO- and paclitaxel-treated Cal51 cells.</title><p>Single-cell copy number profiles for single (<bold>A</bold>) DMSO- and (<bold>B</bold>) paclitaxel-treated cells. A total of 500 Kb genomic bins and DNA content from FACS were used for copy number calculations (see Materials and methods).</p></caption><graphic mime-subtype="jpeg" mimetype="image" xlink:href="elife-69799.xml.media/fig5-figsupp2.jpg"/></fig><fig id="fig5s3" position="float" specific-use="child-fig"><label>Figure 5—figure supplement 3.</label><caption><title>Summary statistic optimization for ABC.</title><p>(<bold>A</bold>) Schematic showing calculation of aneuploidy and MKV. (<bold>B</bold>) Examples of phylogenetic topology metrics. (<bold>C</bold>) Phylogenetic reconstruction of a population of Cal51 cells treated with 20 nM paclitaxel for 48 hr and associated heterogeneity and topology metrics. Normalized and non-normalized summary statistics are displayed (see Materials and methods). (<bold>D</bold>) Analytical scheme to identify most accurate and least variable combinations of heterogeneity and topology metrics. For each combination of 2–9 metrics, we iteratively re-sampled and remeasured the rate of mis-segregation in 100 random cells, three times, from our original dataset of paclitaxel-treated Cal51 cells. The red data point denotes our chosen combination for future analyses—average aneuploidy, MKV, Colless Index, and Cherries. This combination both limits redundant measures (i.e. Colless and Sackin indices) and contains both heterogeneity and topology metrics. (<bold>E</bold>) Percent accuracy and standard error of the mean for three sampled measurements of 100 paclitaxel-treated cells from the original population, repeated for each combination of heterogeneity and topology measures.</p></caption><graphic mime-subtype="jpeg" mimetype="image" xlink:href="elife-69799.xml.media/fig5-figsupp3.jpg"/></fig><fig id="fig5s4" position="float" specific-use="child-fig"><label>Figure 5—figure supplement 4.</label><caption><title>Nullisomy and posterior predictive checks of summary statistics from paclitaxel-treated Cal51 cells.</title><p>(<bold>A</bold>) Observed incidence of nullisomy in paclitaxel-treated cells plotted against the observed mis-segregation rate (P<sub>misseg,true</sub> = 18.5/44 = 0.42) overlaid on simulated data from the second time step (2 generations) under the Hybrid model with S = 0 and P<sub>break</sub> = 0 (n = 3 simulations). (<bold>B</bold>) Posterior distributions of summary statistics from accepted simulations most similar to the paclitaxel-treated Cal51 cells (threshold = 0.05). The red line indicates the observed statistic in paclitaxel-treated cells. Colless index and cherry count is normalized to population size. MKV is normalized to the average ploidy of the population.</p></caption><graphic mime-subtype="jpeg" mimetype="image" xlink:href="elife-69799.xml.media/fig5-figsupp4.jpg"/></fig><fig id="fig5s5" position="float" specific-use="child-fig"><label>Figure 5—figure supplement 5.</label><caption><title>Minimum sampling of karyotype heterogeneity.</title><p>(<bold>A</bold>) Analytical scheme to optimize the number of cells to sample for measuring mis-segregation rates from karyotype heterogeneity. We iteratively re-sampled and remeasured the rate of mis-segregation for a range of sample sizes (n = 5 random samples). (<bold>B</bold>) Predicted mis-segregation rates over a range of sample sizes (n = 5 samples). Points and error bars are the mean ± standard error. Black solid line denotes the mean observed rate of mis-segregation induced by 20 nM paclitaxel. Black dashed lines are half the standard deviation of observed mis-segregation rates per cell. (<bold>C</bold>) Mean percent accuracy of ABC-inferred rates of mis-segregation due to paclitaxel taken from each set of five random samples using the observed rate of mis-segregation as the ‘true value’. Calculated as <inline-formula><mml:math id="inf1"><mml:mstyle displaystyle="true" scriptlevel="0"><mml:mrow><mml:mi mathvariant="normal">M</mml:mi><mml:mi mathvariant="normal">e</mml:mi><mml:mi mathvariant="normal">a</mml:mi><mml:mi mathvariant="normal">n</mml:mi><mml:mspace width="thinmathspace"/><mml:mi mathvariant="normal">%</mml:mi><mml:mspace width="thinmathspace"/><mml:mi mathvariant="normal">a</mml:mi><mml:mi mathvariant="normal">c</mml:mi><mml:mi mathvariant="normal">c</mml:mi><mml:mi mathvariant="normal">u</mml:mi><mml:mi mathvariant="normal">r</mml:mi><mml:mi mathvariant="normal">a</mml:mi><mml:mi mathvariant="normal">c</mml:mi><mml:mi mathvariant="normal">y</mml:mi><mml:mo>=</mml:mo><mml:mn>100</mml:mn><mml:mo>−</mml:mo><mml:mrow><mml:mo>(</mml:mo><mml:mrow><mml:mfrac><mml:mrow><mml:mi mathvariant="normal">t</mml:mi><mml:mi mathvariant="normal">r</mml:mi><mml:mi mathvariant="normal">u</mml:mi><mml:mi mathvariant="normal">e</mml:mi><mml:mo>−</mml:mo><mml:mi mathvariant="normal">m</mml:mi><mml:mi mathvariant="normal">e</mml:mi><mml:mi mathvariant="normal">a</mml:mi><mml:mi mathvariant="normal">n</mml:mi><mml:mspace width="thinmathspace"/><mml:mi mathvariant="normal">i</mml:mi><mml:mi mathvariant="normal">n</mml:mi><mml:mi mathvariant="normal">f</mml:mi><mml:mi mathvariant="normal">e</mml:mi><mml:mi mathvariant="normal">r</mml:mi><mml:mi mathvariant="normal">r</mml:mi><mml:mi mathvariant="normal">e</mml:mi><mml:mi mathvariant="normal">d</mml:mi></mml:mrow><mml:mrow><mml:mi mathvariant="normal">t</mml:mi><mml:mi mathvariant="normal">r</mml:mi><mml:mi mathvariant="normal">u</mml:mi><mml:mi mathvariant="normal">e</mml:mi></mml:mrow></mml:mfrac><mml:mo>×</mml:mo><mml:mn>100</mml:mn></mml:mrow><mml:mo>)</mml:mo></mml:mrow></mml:mrow></mml:mstyle></mml:math></inline-formula>. Dashed lines represent 90% accuracy. (<bold>D</bold>) Standard error of ABC-inferred rates of mis-segregation for each set of random samples from paclitaxel-treated cells. (<bold>E</bold>) ABC-inferred mis-segregation rates by sample size from simulations with known parameters (n = 5 samples). Points represent mean ± standard error across 5 samples for each of 11 selective pressure (S) values. Solid line represents a perfect correlation. Inner dashed line represent ±10% margin. Outer dashed line represents ±20% margin. Simulation parameters: P<sub>misseg</sub>∈ [0, 0.005… 0.02], time steps = 60, Selection Model = ‘Hybrid’, Growth Model = ‘exponential pseudo-Moran’, S = [0, 10... 100], and a tolerance threshold of 0.05. (<bold>F</bold>) Mean percent accuracy of ABC-inferred rates of mis-segregation in simulations (parameters in <bold>E</bold>) taken at various sample sizes. Gray lines represent the mean percent accuracy of five random samples for each sample size for the same simulated population (n = 55 simulations). The dashed line represents 90% accuracy. Calculated as described above but taking the known simulation parameter as the ‘true’ value. (<bold>G</bold>) Standard error of ABC-inferred rates of mis-segregation in simulations (parameters in <bold>E</bold>) taken at various sample sizes. Gray lines represent the standard error of five random samples for each sample size for the same simulated population (n = 55 simulations). (<bold>H</bold>) ABC-inferred mis-segregation rates by sample size from simulations with known parameters (n = 5 samples). Points represent mean ± standard error across 5 samples for each of 11 selective pressure (S) values. Solid line represents a perfect correlation. Inner dashed line represent ±10% margin. Outer dashed line represents ±20% margin. ABC was performed with the following parameters and priors: P<sub>misseg</sub>∈[0, 0.005… 0.05], time steps = 1, Selection Model = ‘Hybrid’, Growth Model = ‘exponential pseudo-Moran’, S ∈ [0, 10… 100], and a tolerance threshold of 0.05. (<bold>I</bold>) Mean percent accuracy of ABC-inferred rates of mis-segregation in simulations (parameters in <bold>H</bold>) taken at various sample sizes. Gray lines represent the mean percent accuracy of five random samples for each sample size for the same simulated population (n = 121 simulations). The dashed line represents 90% accuracy. (<bold>J</bold>) Standard error of ABC-inferred rates of mis-segregation in simulations (parameters in <bold>H</bold>) taken at various sample sizes. Gray lines represent the standard error of five random samples for each sample size for the same simulated population (n = 121 simulations). Note: Red lines in <bold>F</bold>, <bold>G</bold>, <bold>I</bold>, and <bold>J</bold> represent the median.</p></caption><graphic mime-subtype="jpeg" mimetype="image" xlink:href="elife-69799.xml.media/fig5-figsupp5.jpg"/></fig></fig-group><p>In this instance, we were able to estimate mis-segregation rate by calculating absolute deviation from the modal karyotype after a single aberrant cell division. However, such an analysis would not be possible for long-term experiments, or real tumors, where new aneuploid cells may be subject to selection. Accordingly, we sought to infer the parameters of this experiment—the mis-segregation rate of 18.5 chromosomes per division and low selection—using only measures of aneuploidy, variance, and phylogenetic topology. To display this, we used dimensionality reduction to ensure that observed measures from the paclitaxel-treated Cal51 population fell within the space of those observed from simulated populations over 2 steps under the Hybrid model. The experimental data mapped to those from simulations using high mis-segregation rates and relatively low selection (red point, <xref ref-type="fig" rid="fig5">Figure 5D</xref>). However, this comparison does not provide a quantitative measure of CIN. Instead, parameter inference via approximate Bayesian computation (ABC) is well suited for this purpose.</p><p>By deriving phylogeny metrics from simulated populations under a wide-range of distributions of evolutionary parameters, ABC identifies evolutionary parameters most consistent with the data—the posterior probability distribution. We used ABC with simulated data to infer the chromosome mis-segregation rate and selective pressure in the paclitaxel-treated cells (<xref ref-type="bibr" rid="bib14">Csilléry et al., 2012</xref>). Importantly, this data has directly observed rates of mis-segregation, which provide a gold standard benchmark to optimize ABC inference.</p><p>One key aspect of ABC is the selection of optimal phylogenetic summary statistics. A small number of summary statistics is optimal and larger numbers impair the model (<xref ref-type="bibr" rid="bib14">Csilléry et al., 2012</xref>). To address this, a common approach is to identify a small set of summary statistics that achieve the best inference. Here, we used the experimentally observed mis-segregation rate as a benchmark to optimally select a panel of measures for parameter inference (<xref ref-type="fig" rid="fig5s3">Figure 5—figure supplement 3</xref>) and selected the following four metrics to use concurrently in our ABC analysis: mean aneuploidy, MKV, the Colless index (a phylogenetic balance index) and number of cherries (normalized to population size). In doing so, this analysis inferred a chromosome mis-segregation rate of 0.396 ± 0.003 (or 17.4 ± 0.1 chromosomes; mean ± SE), which compares favorably with the experimentally observed rate of 18.5 ± 0.5 (<xref ref-type="fig" rid="fig5">Figure 5E</xref>; dashed line represents experimental rate, white ‘+’ the inferred rate). The distribution of accepted values for selection was skewed toward lower pressure (21 ± 0.4; mean ± SE), meaning that karyotype selection had little bearing on the result at this time point, consistent with the absence of selection in a 48-hr experiment.</p><p>Interestingly, the incidence of nullisomy in the simulated population was higher than in the paclitaxel-treated populations at the observed mis-segregation rate (<xref ref-type="fig" rid="fig5s4">Figure 5—figure supplement 4A</xref>). This could be due to spindle pole clustering, a recovery mechanism often seen in paclitaxel-treated cells that causes non-random chromosome mis-segregations. A posterior predictive check of the summary statistics demonstrates how each contributes to the inference of CIN rate (<xref ref-type="fig" rid="fig5s4">Figure 5—figure supplement 4B</xref>). In short, this experimental case validated ABC-derived mis-segregation rate as a measure of CIN, with an experimentally determined mis-segregation rate. Importantly, prior estimations of mis-segregation rate selective pressure were not required to develop this quantitative measure of CIN.</p><p>Together, these data indicate that combining simulated and observed metrics of population diversity and structure with a Bayesian framework for parameter inference is a flexible method of quantifying the evolutionary forces associated with CIN. Moreover, this method reveals the hitherto unreported potential extent of chromosome mis-segregation induced by a clinically relevant concentration of the successful chemotherapeutic paclitaxel consistent with the measured mis-segregation from non-pharmacologically induced multipolar divisions (<xref ref-type="bibr" rid="bib9">Bollen et al., 2021</xref>).</p></sec><sec id="s2-6"><title>Minimum sampling of karyotype heterogeneity</title><p>The cost of high-throughput DNA sequencing of single cells is often cited as a limitation to clinical implementation (<xref ref-type="bibr" rid="bib23">Evrony et al., 2021</xref>). In part, the cost can be limited by low-coverage sequencing which is sufficient to estimate the density of reads across the genome. Further, it may be possible to minimize the number of cells that are sampled to get a robust estimate of CIN, though sampling too few cells may result in inaccurate measurements. Accordingly, we determined how sampling impacts measurement of mis-segregation rates using approximate Bayesian computation. We first took five random samples from the population of paclitaxel-treated cells each at various sample sizes (<xref ref-type="fig" rid="fig5s5">Figure 5—figure supplement 5A</xref>). We then inferred the mis-segregation rate in each sample and identified the sample size that surpasses an average of 90% accuracy and a low standard error of measurement. We found that even small sample sizes can accurately infer the mis-segregation rate, in this context, with a low standard error (<xref ref-type="fig" rid="fig5s5">Figure 5—figure supplement 5B-D</xref>). A sample size of 60 cells produced the most accurate measurement at 99.5% and a standard error of 0.008 ( ± 0.35 chromosomes). We repeated this analysis using simulated data from the Hybrid selection model and a range of mis-segregation rates spanning what is observed in cancer and non-cancer cultures (P<sub>misseg</sub> ≤ 0.02; see below). We again found a range of sample sizes whose inferred mis-segregation rates underestimate the known value from those simulations (n∈ [20, 40… 180]; <xref ref-type="fig" rid="fig5s5">Figure 5—figure supplement 5E</xref>,F). Across all mis-segregation rates and selective pressures, random samples of 200 cells had a median percent accuracy of 90% and median standard error of 0.0003 ( ± 0.0138 chromosomes per division). The difference in optimal sample sizes between the paclitaxel-treated population and the simulated population is notable and likely due to the presence of ‘clonal’ structures in the simulated population. While the paclitaxel treatment resulted in a uniformly high degree of aneuploidy and little evidence of karyotype selection, the simulated populations after 60 steps (~30 generations) have discrete copy number clusters that may not be captured in each random sample. To verify this, we repeated the analysis using only data from the first time step, prior to the onset of karyotype selection (<xref ref-type="fig" rid="fig5s5">Figure 5—figure supplement 5H</xref>). In this case, we found that the sample size needed to achieve a median 90% accuracy over all simulations in this context is 100 cells, at which point the standard error for P<sub>misseg</sub> is 0.0068 (placing measures within ±0.31 chromosomes per division; <xref ref-type="fig" rid="fig5s5">Figure 5—figure supplement 5I, J</xref>). Thus, a larger number of cells is required in the context of long-term karyotype selection than a more acute time scale, such as we see with paclitaxel.</p><p>In conclusion, we recommend using 200 cells from a single sampled site which, at biologically relevant time scales and rates of mis-segregation, provides ≥90% accuracy. These data represent, to our knowledge, the first analysis of how sample size for single-cell sequencing affects the accuracy and measurement of chromosome mis-segregation rates.</p></sec><sec id="s2-7"><title>Inferring chromosome mis-segregation rates in tumors and organoids</title><p>To determine if this framework is clinically applicable, we employed previously published scDNAseq datasets derived from tumor samples and patient-derived organoids (PDO) (<xref ref-type="bibr" rid="bib8">Bolhaqueiro et al., 2019</xref>; <xref ref-type="bibr" rid="bib52">Navin et al., 2011</xref>). Importantly, the data from Bolhaqueiro et al. include sample-matched live cell imaging data in colorectal cancer PDOs, with direct observation of chromosome mis-segregation events to compare with inferred measures. We established our panel of measurements on these populations (<xref ref-type="fig" rid="fig6">Figure 6A</xref>) and used these to tune the prior distribution of time steps and the rejection threshold for ABC. In sensitivity analysis, 20 steps or greater was sufficient to establish stable estimates of P<sub>misseg</sub> and selection, S (<xref ref-type="fig" rid="fig6s1">Figure 6—figure supplement 1A-B</xref>)—we chose a window of 40–80 steps for further analysis. For rejection thresholds 0.05 and smaller, the inferred mis-segregation rates remained steady (<xref ref-type="fig" rid="fig6s1">Figure 6—figure supplement 1C</xref>). With these model parameters chosen, we evaluated the different selection models, and found that the Abundance model resulted in simulated data that best resembled experimental data, for both exponential and constant-population dynamics (<xref ref-type="table" rid="table3">Table 3</xref>). Given that the Abundance model is the most biologically relevant, we will use data simulated under this model in our prior dataset for inference.</p><fig-group><fig id="fig6" position="float"><label>Figure 6.</label><caption><title>Inferring chromosome mis-segregation rates in tumors and organoids <xref ref-type="bibr" rid="bib8">Bolhaqueiro et al., 2019</xref><xref ref-type="bibr" rid="bib52">Navin et al., 2011</xref>.</title><p>(<bold>A</bold>) Computed population summary statistics for colorectal cancer (CRC) patient-derived organoids (PDOs) and breast biopsy scDNAseq datasets from <xref ref-type="bibr" rid="bib8">Bolhaqueiro et al., 2019</xref> (gold) and <xref ref-type="bibr" rid="bib52">Navin et al., 2011</xref> (pink). (<bold>B</bold>) Dimensionality reduction analysis of population summary statistics showing biological observations overlaid on, and found within, the space of simulated observations. Point colors show the simulation parameters and summary statistics for all simulations using the following priors and parameters: Growth Model = ‘exponential pseudo-Moran’, Selection Model = ‘Abundance’, initial ploidy = 2, time steps ∈[40, 41… 80], S ∈[0,2… 100], P<sub>misseg</sub>∈[0,0.001… 0.050] and a tolerance threshold of 0.05 to reject dissimilar simulation results. (see Materials and Methods). (<bold>C</bold>) 2D density plots showing joint posterior distributions of P<sub>misseg</sub> and S values from the approximate Bayesian computation analysis of samples 26 N (left) and 24Tb (right) from <xref ref-type="bibr" rid="bib8">Bolhaqueiro et al., 2019</xref>. White + represents the mean of inferred values. (<bold>D</bold>) Inferred selective pressures and mis-segregation rates from each scDNAseq dataset (mean and SEM of accepted values). (<bold>E</bold>) Predicted mis-segregation rates in CRC PDOs and a breast biopsy plotted with approximated mis-segregation rates observed in cancer (blue triangle) and non-cancer (red circle) models (primarily cell lines) from previous studies (<xref ref-type="table" rid="table5">Table 5</xref>; see Materials and methods). The predicted mis-segregation rates in these cancer-derived samples fall within those observed in cancer cell lines and above those of non-cancer cell lines. (<bold>F</bold>) Pearson correlation of predicted mis-segregation rates and predicted selective pressures in CRC PDOs from <xref ref-type="bibr" rid="bib8">Bolhaqueiro et al., 2019</xref>. (<bold>G</bold>) Pearson correlation of predicted mis-segregation rates and the incidence of observed segregation errors in CRC PDOs from <xref ref-type="bibr" rid="bib8">Bolhaqueiro et al., 2019</xref>. Error bars represent SEM values. (<bold>H</bold>) Pearson correlation of observed incidence of segregation errors in CRC PDOs from <xref ref-type="bibr" rid="bib8">Bolhaqueiro et al., 2019</xref> to the ploidy-corrected prediction of the observed incidence of segregation errors. These values assume the involvement of 1 chromosome per observed error and are calculated as the (predicted mis-segregation rate) x (mean number of chromosomes observed per cell) x 100. Dotted line = 1:1 reference.</p></caption><graphic mime-subtype="jpeg" mimetype="image" xlink:href="elife-69799.xml.media/fig6.jpg"/></fig><fig id="fig6s1" position="float" specific-use="child-fig"><label>Figure 6—figure supplement 1.</label><caption><title>ABC-inference threshold and step-window analysis.</title><p>Posterior distributions of mis-segregation rates (<bold>A</bold>) and selective pressure, S (<bold>B</bold>) inferred using ABC analysis of CRC organoids and a breast biopsy from <xref ref-type="bibr" rid="bib8">Bolhaqueiro et al., 2019</xref> and <xref ref-type="bibr" rid="bib52">Navin et al., 2011</xref> respectively using a sliding window prior distribution of time steps. ABC was performed for every interval of 10 steps between 0 and 100 using a tolerance threshold of 0.05. Schematic of analysis shown below. ABC was performed with the following parameters and priors: P<sub>misseg</sub>∈ [0...0.001...0.05], S ∈ [0...2...100], indicated time step window, Selection Model = ‘Abundance’, Growth Model = ‘exponential pseudo-Moran’, and a tolerance threshold of 0.05. (<bold>C</bold>) Posterior distributions of mis-segregation rates inferred using ABC analysis on the same samples as in A using tolerance thresholds of 0.005, 0.01, 0.05, 0.1. ABC was performed with the following parameters and priors: P<sub>misseg</sub>∈ [0, 0.001… 0.05], S ∈ [0, 2… 100], time steps ∈ [40, 41… 80], Selection Model = ‘Abundance’, Growth Model = ‘exponential pseudo-Moran’, and the indicated tolerance threshold.</p></caption><graphic mime-subtype="jpeg" mimetype="image" xlink:href="elife-69799.xml.media/fig6-figsupp1.jpg"/></fig><fig id="fig6s2" position="float" specific-use="child-fig"><label>Figure 6—figure supplement 2.</label><caption><title>ABC-inferred step count in patient-derived samples.</title><p>Mean and standard error for steps in each patient-derived sample (accompanying data in <xref ref-type="fig" rid="fig6">Figure 6</xref>), inferred via approximate Bayesian computation.</p></caption><graphic mime-subtype="jpeg" mimetype="image" xlink:href="elife-69799.xml.media/fig6-figsupp2.jpg"/></fig><fig id="fig6s3" position="float" specific-use="child-fig"><label>Figure 6—figure supplement 3.</label><caption><title>ABC-inferred mis-segregation rates and selective pressures in patient-derived samples.</title><p>Joint (2D density plots) and individual (1D density plots) distributions of mis-segregation rates and selective pressures in patient-derived CRC organoids and a breast biopsy from <xref ref-type="bibr" rid="bib8">Bolhaqueiro et al., 2019</xref> and <xref ref-type="bibr" rid="bib52">Navin et al., 2011</xref> respectively (accompanying data in <xref ref-type="fig" rid="fig6">Figure 6</xref>). The prior (yellow) distribution represents the parameters used for simulation while the posterior (gray) distribution represents the parameters from simulations whose observed measurements were similar to the measurements taken from the patient-derived sample using a tolerance threshold of 0.05. White + signs on joint distributions represent the mean of both parameters.</p></caption><graphic mime-subtype="jpeg" mimetype="image" xlink:href="elife-69799.xml.media/fig6-figsupp3.jpg"/></fig><fig id="fig6s4" position="float" specific-use="child-fig"><label>Figure 6—figure supplement 4.</label><caption><title>Validation of selection in longitudinally sequenced CRC organoids.</title><p>(<bold>A–C</bold>) Copy number heatmaps showing the deviation from the mode of each chromosome derived from longitudinally sequenced clonal organoids from <xref ref-type="bibr" rid="bib8">Bolhaqueiro et al., 2019</xref>. ABC was performed on scDNAseq data from three clones at 3 weeks of growth. The resulting inferred mis-segregation rate (P<sub>misseg</sub>) and selective pressure (S) were used to simulate CIN and selection in these clones over 60 time steps, at which point the composition of the populations were compared to the scDNAseq data from each of the clones at 24 weeks of growth (<bold>D–K</bold>). Additional simulations using S = 0 (not shown) and S = 1 were also performed. Inferred P<sub>misseg</sub> values for (<bold>A</bold>) clone 1, (<bold>B</bold>) clone 2, and (<bold>C</bold>) clone 3 were 0.0042, 0.0046, and 0.0051 respectively. S = 60 was inferred for each clone. ABC was performed on the 3 week data with the following parameters and priors: P<sub>misseg</sub>∈ [0, 0.001... 0.05], S ∈ [0, 2… 100], time steps ∈ [40, 41... 80], Selection Model = ‘Abundance’, Growth Model = ‘exponential pseudo-Moran’, and a tolerance threshold of 0.05. (<bold>D</bold>) MKV values from n = 10 simulations per clone. Dotted line represents the MKV value observed in the scDNAseq data. (<bold>E</bold>) Aneuploidy values from n = 10 simulations per clone per S value. Dotted line represents the Aneuploidy value observed in the scDNAseq data. (<bold>F</bold>) Colless index values from n = 10 simulations per clone S value. Dotted line represents the Colless index value observed in the scDNAseq data. (<bold>G</bold>) Normalized cherry values from n = 10 simulations per clone S value. Dotted line represents the normalized cherry value observed in the scDNAseq data. (<bold>H</bold>) Percent error for MKV observations in n = 10 simulations per clone per S value. Dotted line represents 0% error. (<bold>I</bold>) Percent error for aneuploidy observations in n = 10 simulations per clone per S value. Dotted line represents 0% error. (<bold>J</bold>) Percent error for Colless observations in n = 10 simulations per clone per S value. Dotted line represents 0% error. (<bold>K</bold>) Percent error for normalized cherry observations in n = 10 simulations per clone per S value. Dotted line represents 0% error.</p></caption><graphic mime-subtype="jpeg" mimetype="image" xlink:href="elife-69799.xml.media/fig6-figsupp4.jpg"/></fig><fig id="fig6s5" position="float" specific-use="child-fig"><label>Figure 6—figure supplement 5.</label><caption><title>Joint posterior distributions from CRC organoids at 3 weeks.</title><p>Joint (2D density plots) and individual (1D density plots) distributions of mis-segregation rates and selective pressures in individual clones of a patient-derived CRC organoid line from <xref ref-type="bibr" rid="bib8">Bolhaqueiro et al., 2019</xref> after 3 weeks of growth (accompanying data in <xref ref-type="fig" rid="fig6s4">Figure 6—figure supplement 4</xref>). The prior (yellow) distribution represents the parameters used for simulation while the posterior (gray) distribution represents the parameters from simulations whose observed measurements were similar to the measurements taken from the patient-derived sample using a tolerance threshold of 0.05. White + signs on joint distributions represent the mean of both parameters.</p></caption><graphic mime-subtype="jpeg" mimetype="image" xlink:href="elife-69799.xml.media/fig6-figsupp5.jpg"/></fig></fig-group><table-wrap id="table3" position="float"><label>Table 3.</label><caption><title>Model selection.</title></caption><table frame="hsides" rules="groups"><thead><tr><th align="left" valign="bottom">Sample</th><th align="left" valign="bottom">Growt Model</th><th align="left" valign="bottom">Selectio Model</th><th align="left" valign="bottom">PP</th><th align="left" valign="bottom">BF (Ho Neutral)</th><th align="left" valign="bottom">Pmisseg</th><th align="left" valign="bottom">S</th><th align="left" valign="bottom">Steps</th></tr></thead><tbody><tr><td align="left" valign="bottom">7T</td><td align="left" valign="bottom">exponential pseudo-Moran</td><td align="left" valign="bottom">Abundance</td><td align="left" valign="bottom">0.621</td><td align="left" valign="bottom">Inf</td><td align="left" valign="bottom">0.0033 ± 1e-05</td><td align="left" valign="bottom">60.5416 ± 0.2053</td><td align="left" valign="bottom">59.8475 ± 0.0937</td></tr><tr><td align="left" valign="bottom">7T</td><td align="left" valign="bottom">exponential pseudo-Moran</td><td align="left" valign="bottom">Driver</td><td align="left" valign="bottom">0.14</td><td align="left" valign="bottom">Inf</td><td align="left" valign="bottom">0.001 ± 1e-05</td><td align="left" valign="bottom">49.6557 ± 0.2389</td><td align="left" valign="bottom">58.7002 ± 0.0943</td></tr><tr><td align="left" valign="bottom">7T</td><td align="left" valign="bottom">exponential pseudo-Moran</td><td align="left" valign="bottom">Hybrid</td><td align="left" valign="bottom">0.239</td><td align="left" valign="bottom">Inf</td><td align="left" valign="bottom">8e-04 ± 1e-05</td><td align="left" valign="bottom">49.3428 ± 0.2377</td><td align="left" valign="bottom">58.5789 ± 0.0935</td></tr><tr><td align="left" valign="bottom">7T</td><td align="left" valign="bottom">exponential pseudo-Moran</td><td align="left" valign="bottom">Neutral</td><td align="left" valign="bottom">0</td><td align="left" valign="bottom">NA</td><td align="left" valign="bottom">9e-04 ± 5e-05</td><td align="left" valign="bottom">0 ± 0</td><td align="left" valign="bottom">57.7994 ± 0.6728</td></tr><tr><td align="left" valign="bottom">7T</td><td align="left" valign="bottom">constant Wright-Fisher</td><td align="left" valign="bottom">Abundance</td><td align="left" valign="bottom">0.985</td><td align="left" valign="bottom">Inf</td><td align="left" valign="bottom">0.0062 ± 2e-05</td><td align="left" valign="bottom">69.7026 ± 0.1724</td><td align="left" valign="bottom">59.9318 ± 0.0937</td></tr><tr><td align="left" valign="bottom">7T</td><td align="left" valign="bottom">constant Wright-Fisher</td><td align="left" valign="bottom">Driver</td><td align="left" valign="bottom">0</td><td align="left" valign="bottom">NA</td><td align="left" valign="bottom">0.0012 ± 1e-05</td><td align="left" valign="bottom">48.2881 ± 0.2384</td><td align="left" valign="bottom">57.5239 ± 0.0933</td></tr><tr><td align="left" valign="bottom">7T</td><td align="left" valign="bottom">constant Wright-Fisher</td><td align="left" valign="bottom">Hybrid</td><td align="left" valign="bottom">0.015</td><td align="left" valign="bottom">Inf</td><td align="left" valign="bottom">9e-04 ± 1e-05</td><td align="left" valign="bottom">50.7803 ± 0.2359</td><td align="left" valign="bottom">58.2514 ± 0.0941</td></tr><tr><td align="left" valign="bottom">7T</td><td align="left" valign="bottom">constant Wright-Fisher</td><td align="left" valign="bottom">Neutral</td><td align="left" valign="bottom">0</td><td align="left" valign="bottom">NA</td><td align="left" valign="bottom">9e-04 ± 5e-05</td><td align="left" valign="bottom">0 ± 0</td><td align="left" valign="bottom">58.7803 ± 0.6701</td></tr><tr><td align="left" valign="bottom">U1T</td><td align="left" valign="bottom">exponential pseudo-Moran</td><td align="left" valign="bottom">Abundance</td><td align="left" valign="bottom">0.582</td><td align="left" valign="bottom">199</td><td align="left" valign="bottom">9e-04 ± 1e-05</td><td align="left" valign="bottom">56.8672 ± 0.2168</td><td align="left" valign="bottom">59.9906 ± 0.0937</td></tr><tr><td align="left" valign="bottom">U1T</td><td align="left" valign="bottom">exponential pseudo-Moran</td><td align="left" valign="bottom">Driver</td><td align="left" valign="bottom">0.113</td><td align="left" valign="bottom">39</td><td align="left" valign="bottom">0.001 ± 1e-05</td><td align="left" valign="bottom">49.6611 ± 0.2389</td><td align="left" valign="bottom">58.6886 ± 0.0944</td></tr><tr><td align="left" valign="bottom">U1T</td><td align="left" valign="bottom">exponential pseudo-Moran</td><td align="left" valign="bottom">Hybrid</td><td align="left" valign="bottom">0.156</td><td align="left" valign="bottom">54</td><td align="left" valign="bottom">8e-04 ± 1e-05</td><td align="left" valign="bottom">49.3658 ± 0.2375</td><td align="left" valign="bottom">58.569 ± 0.0935</td></tr><tr><td align="left" valign="bottom">U1T</td><td align="left" valign="bottom">exponential pseudo-Moran</td><td align="left" valign="bottom">Neutral</td><td align="left" valign="bottom">0.149</td><td align="left" valign="bottom">1</td><td align="left" valign="bottom">9e-04 ± 5e-05</td><td align="left" valign="bottom">0 ± 0</td><td align="left" valign="bottom">57.7102 ± 0.67</td></tr><tr><td align="left" valign="bottom">U1T</td><td align="left" valign="bottom">constant Wright-Fisher</td><td align="left" valign="bottom">Abundance</td><td align="left" valign="bottom">0.654</td><td align="left" valign="bottom">290</td><td align="left" valign="bottom">0.001 ± 1e-05</td><td align="left" valign="bottom">61.4358 ± 0.2029</td><td align="left" valign="bottom">60.0021 ± 0.0937</td></tr><tr><td align="left" valign="bottom">U1T</td><td align="left" valign="bottom">constant Wright-Fisher</td><td align="left" valign="bottom">Driver</td><td align="left" valign="bottom">0.115</td><td align="left" valign="bottom">51</td><td align="left" valign="bottom">0.0012 ± 1e-05</td><td align="left" valign="bottom">48.2767 ± 0.2383</td><td align="left" valign="bottom">57.5267 ± 0.0934</td></tr><tr><td align="left" valign="bottom">U1T</td><td align="left" valign="bottom">constant Wright-Fisher</td><td align="left" valign="bottom">Hybrid</td><td align="left" valign="bottom">0.115</td><td align="left" valign="bottom">51</td><td align="left" valign="bottom">9e-04 ± 1e-05</td><td align="left" valign="bottom">50.8033 ± 0.2358</td><td align="left" valign="bottom">58.2507 ± 0.0941</td></tr><tr><td align="left" valign="bottom">U1T</td><td align="left" valign="bottom">constant Wright-Fisher</td><td align="left" valign="bottom">Neutral</td><td align="left" valign="bottom">0.115</td><td align="left" valign="bottom">1</td><td align="left" valign="bottom">9e-04 ± 5e-05</td><td align="left" valign="bottom">0 ± 0</td><td align="left" valign="bottom">58.7803 ± 0.6701</td></tr><tr><td align="left" valign="bottom">U2T</td><td align="left" valign="bottom">exponential pseudo-Moran</td><td align="left" valign="bottom">Abundance</td><td align="left" valign="bottom">0.628</td><td align="left" valign="bottom">251</td><td align="left" valign="bottom">0.0054 ± 1e-05</td><td align="left" valign="bottom">59.4269 ± 0.2108</td><td align="left" valign="bottom">59.8349 ± 0.0935</td></tr><tr><td align="left" valign="bottom">U2T</td><td align="left" valign="bottom">exponential pseudo-Moran</td><td align="left" valign="bottom">Driver</td><td align="left" valign="bottom">0.079</td><td align="left" valign="bottom">32</td><td align="left" valign="bottom">0.0027 ± 2e-05</td><td align="left" valign="bottom">50.1513 ± 0.2396</td><td align="left" valign="bottom">57.4538 ± 0.0934</td></tr><tr><td align="left" valign="bottom">U2T</td><td align="left" valign="bottom">exponential pseudo-Moran</td><td align="left" valign="bottom">Hybrid</td><td align="left" valign="bottom">0.166</td><td align="left" valign="bottom">66</td><td align="left" valign="bottom">0.0022 ± 2e-05</td><td align="left" valign="bottom">48.7779 ± 0.2413</td><td align="left" valign="bottom">57.7078 ± 0.0934</td></tr><tr><td align="left" valign="bottom">U2T</td><td align="left" valign="bottom">exponential pseudo-Moran</td><td align="left" valign="bottom">Neutral</td><td align="left" valign="bottom">0.127</td><td align="left" valign="bottom">1</td><td align="left" valign="bottom">0.0021 ± 7e-05</td><td align="left" valign="bottom">0 ± 0</td><td align="left" valign="bottom">56.8535 ± 0.6619</td></tr><tr><td align="left" valign="bottom">U2T</td><td align="left" valign="bottom">constant Wright-Fisher</td><td align="left" valign="bottom">Abundance</td><td align="left" valign="bottom">0.918</td><td align="left" valign="bottom">2817</td><td align="left" valign="bottom">0.0112 ± 3e-05</td><td align="left" valign="bottom">69.7222 ± 0.1703</td><td align="left" valign="bottom">60.0655 ± 0.0934</td></tr><tr><td align="left" valign="bottom">U2T</td><td align="left" valign="bottom">constant Wright-Fisher</td><td align="left" valign="bottom">Driver</td><td align="left" valign="bottom">0.001</td><td align="left" valign="bottom">4</td><td align="left" valign="bottom">0.0027 ± 2e-05</td><td align="left" valign="bottom">48.7794 ± 0.2389</td><td align="left" valign="bottom">56.4812 ± 0.0919</td></tr><tr><td align="left" valign="bottom">U2T</td><td align="left" valign="bottom">constant Wright-Fisher</td><td align="left" valign="bottom">Hybrid</td><td align="left" valign="bottom">0.064</td><td align="left" valign="bottom">196</td><td align="left" valign="bottom">0.0022 ± 1e-05</td><td align="left" valign="bottom">50.9564 ± 0.2379</td><td align="left" valign="bottom">57.1161 ± 0.0925</td></tr><tr><td align="left" valign="bottom">U2T</td><td align="left" valign="bottom">constant Wright-Fisher</td><td align="left" valign="bottom">Neutral</td><td align="left" valign="bottom">0.017</td><td align="left" valign="bottom">1</td><td align="left" valign="bottom">0.0022 ± 1e-04</td><td align="left" valign="bottom">0 ± 0</td><td align="left" valign="bottom">57.7898 ± 0.6841</td></tr><tr><td align="left" valign="bottom">U3T</td><td align="left" valign="bottom">exponential pseudo-Moran</td><td align="left" valign="bottom">Abundance</td><td align="left" valign="bottom">0.582</td><td align="left" valign="bottom">199</td><td align="left" valign="bottom">0.0029 ± 1e-05</td><td align="left" valign="bottom">60.9557 ± 0.2091</td><td align="left" valign="bottom">59.8273 ± 0.0938</td></tr><tr><td align="left" valign="bottom">U3T</td><td align="left" valign="bottom">exponential pseudo-Moran</td><td align="left" valign="bottom">Driver</td><td align="left" valign="bottom">0.113</td><td align="left" valign="bottom">39</td><td align="left" valign="bottom">0.001 ± 1e-05</td><td align="left" valign="bottom">49.6707 ± 0.2389</td><td align="left" valign="bottom">58.6986 ± 0.0944</td></tr><tr><td align="left" valign="bottom">U3T</td><td align="left" valign="bottom">exponential pseudo-Moran</td><td align="left" valign="bottom">Hybrid</td><td align="left" valign="bottom">0.156</td><td align="left" valign="bottom">54</td><td align="left" valign="bottom">8e-04 ± 1e-05</td><td align="left" valign="bottom">49.3754 ± 0.2376</td><td align="left" valign="bottom">58.5711 ± 0.0935</td></tr><tr><td align="left" valign="bottom">U3T</td><td align="left" valign="bottom">exponential pseudo-Moran</td><td align="left" valign="bottom">Neutral</td><td align="left" valign="bottom">0.149</td><td align="left" valign="bottom">1</td><td align="left" valign="bottom">9e-04 ± 5e-05</td><td align="left" valign="bottom">0 ± 0</td><td align="left" valign="bottom">57.7102 ± 0.67</td></tr><tr><td align="left" valign="bottom">U3T</td><td align="left" valign="bottom">constant Wright-Fisher</td><td align="left" valign="bottom">Abundance</td><td align="left" valign="bottom">0.736</td><td align="left" valign="bottom">Inf</td><td align="left" valign="bottom">0.0052 ± 2e-05</td><td align="left" valign="bottom">69.8357 ± 0.1713</td><td align="left" valign="bottom">59.932 ± 0.0934</td></tr><tr><td align="left" valign="bottom">U3T</td><td align="left" valign="bottom">constant Wright-Fisher</td><td align="left" valign="bottom">Driver</td><td align="left" valign="bottom">0.13</td><td align="left" valign="bottom">Inf</td><td align="left" valign="bottom">0.0012 ± 1e-05</td><td align="left" valign="bottom">48.2864 ± 0.2383</td><td align="left" valign="bottom">57.5385 ± 0.0934</td></tr><tr><td align="left" valign="bottom">U3T</td><td align="left" valign="bottom">constant Wright-Fisher</td><td align="left" valign="bottom">Hybrid</td><td align="left" valign="bottom">0.134</td><td align="left" valign="bottom">Inf</td><td align="left" valign="bottom">9e-04 ± 1e-05</td><td align="left" valign="bottom">50.8219 ± 0.2357</td><td align="left" valign="bottom">58.2482 ± 0.0941</td></tr><tr><td align="left" valign="bottom">U3T</td><td align="left" valign="bottom">constant Wright-Fisher</td><td align="left" valign="bottom">Neutral</td><td align="left" valign="bottom">0</td><td align="left" valign="bottom">NA</td><td align="left" valign="bottom">9e-04 ± 5e-05</td><td align="left" valign="bottom">0 ± 0</td><td align="left" valign="bottom">58.8567 ± 0.6676</td></tr><tr><td align="left" valign="bottom">14T</td><td align="left" valign="bottom">exponential pseudo-Moran</td><td align="left" valign="bottom">Abundance</td><td align="left" valign="bottom">0.582</td><td align="left" valign="bottom">199</td><td align="left" valign="bottom">9e-04 ± 1e-05</td><td align="left" valign="bottom">56.8672 ± 0.2168</td><td align="left" valign="bottom">59.9906 ± 0.0937</td></tr><tr><td align="left" valign="bottom">14T</td><td align="left" valign="bottom">exponential pseudo-Moran</td><td align="left" valign="bottom">Driver</td><td align="left" valign="bottom">0.113</td><td align="left" valign="bottom">39</td><td align="left" valign="bottom">0.001 ± 1e-05</td><td align="left" valign="bottom">49.6614 ± 0.239</td><td align="left" valign="bottom">58.695 ± 0.0944</td></tr><tr><td align="left" valign="bottom">14T</td><td align="left" valign="bottom">exponential pseudo-Moran</td><td align="left" valign="bottom">Hybrid</td><td align="left" valign="bottom">0.156</td><td align="left" valign="bottom">54</td><td align="left" valign="bottom">8e-04 ± 1e-05</td><td align="left" valign="bottom">49.3716 ± 0.2375</td><td align="left" valign="bottom">58.5632 ± 0.0935</td></tr><tr><td align="left" valign="bottom">14T</td><td align="left" valign="bottom">exponential pseudo-Moran</td><td align="left" valign="bottom">Neutral</td><td align="left" valign="bottom">0.149</td><td align="left" valign="bottom">1</td><td align="left" valign="bottom">9e-04 ± 5e-05</td><td align="left" valign="bottom">0 ± 0</td><td align="left" valign="bottom">57.7102 ± 0.67</td></tr><tr><td align="left" valign="bottom">14T</td><td align="left" valign="bottom">constant Wright-Fisher</td><td align="left" valign="bottom">Abundance</td><td align="left" valign="bottom">0.654</td><td align="left" valign="bottom">290</td><td align="left" valign="bottom">0.0011 ± 1e-05</td><td align="left" valign="bottom">62.8579 ± 0.2075</td><td align="left" valign="bottom">60.0029 ± 0.0936</td></tr><tr><td align="left" valign="bottom">14T</td><td align="left" valign="bottom">constant Wright-Fisher</td><td align="left" valign="bottom">Driver</td><td align="left" valign="bottom">0.115</td><td align="left" valign="bottom">51</td><td align="left" valign="bottom">0.0012 ± 1e-05</td><td align="left" valign="bottom">48.2967 ± 0.2383</td><td align="left" valign="bottom">57.5295 ± 0.0934</td></tr><tr><td align="left" valign="bottom">14T</td><td align="left" valign="bottom">constant Wright-Fisher</td><td align="left" valign="bottom">Hybrid</td><td align="left" valign="bottom">0.115</td><td align="left" valign="bottom">51</td><td align="left" valign="bottom">9e-04 ± 1e-05</td><td align="left" valign="bottom">50.8274 ± 0.2357</td><td align="left" valign="bottom">58.2478 ± 0.0941</td></tr><tr><td align="left" valign="bottom">14T</td><td align="left" valign="bottom">constant Wright-Fisher</td><td align="left" valign="bottom">Neutral</td><td align="left" valign="bottom">0.115</td><td align="left" valign="bottom">1</td><td align="left" valign="bottom">9e-04 ± 5e-05</td><td align="left" valign="bottom">0 ± 0</td><td align="left" valign="bottom">58.8567 ± 0.6676</td></tr><tr><td align="left" valign="bottom">16T</td><td align="left" valign="bottom">exponential pseudo-Moran</td><td align="left" valign="bottom">Abundance</td><td align="left" valign="bottom">0.582</td><td align="left" valign="bottom">199</td><td align="left" valign="bottom">0.002 ± 1e-05</td><td align="left" valign="bottom">61.2401 ± 0.2028</td><td align="left" valign="bottom">59.9109 ± 0.0935</td></tr><tr><td align="left" valign="bottom">16T</td><td align="left" valign="bottom">exponential pseudo-Moran</td><td align="left" valign="bottom">Driver</td><td align="left" valign="bottom">0.113</td><td align="left" valign="bottom">39</td><td align="left" valign="bottom">0.001 ± 1e-05</td><td align="left" valign="bottom">49.6539 ± 0.2389</td><td align="left" valign="bottom">58.7006 ± 0.0943</td></tr><tr><td align="left" valign="bottom">16T</td><td align="left" valign="bottom">exponential pseudo-Moran</td><td align="left" valign="bottom">Hybrid</td><td align="left" valign="bottom">0.156</td><td align="left" valign="bottom">54</td><td align="left" valign="bottom">8e-04 ± 1e-05</td><td align="left" valign="bottom">49.3611 ± 0.2376</td><td align="left" valign="bottom">58.574 ± 0.0935</td></tr><tr><td align="left" valign="bottom">16T</td><td align="left" valign="bottom">exponential pseudo-Moran</td><td align="left" valign="bottom">Neutral</td><td align="left" valign="bottom">0.149</td><td align="left" valign="bottom">1</td><td align="left" valign="bottom">9e-04 ± 5e-05</td><td align="left" valign="bottom">0 ± 0</td><td align="left" valign="bottom">57.7994 ± 0.6728</td></tr><tr><td align="left" valign="bottom">16T</td><td align="left" valign="bottom">constant Wright-Fisher</td><td align="left" valign="bottom">Abundance</td><td align="left" valign="bottom">0.654</td><td align="left" valign="bottom">290</td><td align="left" valign="bottom">0.0038 ± 1e-05</td><td align="left" valign="bottom">69.8456 ± 0.1701</td><td align="left" valign="bottom">59.9523 ± 0.0936</td></tr><tr><td align="left" valign="bottom">16T</td><td align="left" valign="bottom">constant Wright-Fisher</td><td align="left" valign="bottom">Driver</td><td align="left" valign="bottom">0.115</td><td align="left" valign="bottom">51</td><td align="left" valign="bottom">0.0012 ± 1e-05</td><td align="left" valign="bottom">48.261 ± 0.2384</td><td align="left" valign="bottom">57.5233 ± 0.0933</td></tr><tr><td align="left" valign="bottom">16T</td><td align="left" valign="bottom">constant Wright-Fisher</td><td align="left" valign="bottom">Hybrid</td><td align="left" valign="bottom">0.115</td><td align="left" valign="bottom">51</td><td align="left" valign="bottom">9e-04 ± 1e-05</td><td align="left" valign="bottom">50.7713 ± 0.2359</td><td align="left" valign="bottom">58.2554 ± 0.0941</td></tr><tr><td align="left" valign="bottom">16T</td><td align="left" valign="bottom">constant Wright-Fisher</td><td align="left" valign="bottom">Neutral</td><td align="left" valign="bottom">0.115</td><td align="left" valign="bottom">1</td><td align="left" valign="bottom">9e-04 ± 5e-05</td><td align="left" valign="bottom">0 ± 0</td><td align="left" valign="bottom">58.7803 ± 0.6701</td></tr><tr><td align="left" valign="bottom">19Ta</td><td align="left" valign="bottom">exponential pseudo-Moran</td><td align="left" valign="bottom">Abundance</td><td align="left" valign="bottom">0.711</td><td align="left" valign="bottom">313</td><td align="left" valign="bottom">0.004 ± 1e-05</td><td align="left" valign="bottom">60.6391 ± 0.2074</td><td align="left" valign="bottom">59.7801 ± 0.0934</td></tr><tr><td align="left" valign="bottom">19Ta</td><td align="left" valign="bottom">exponential pseudo-Moran</td><td align="left" valign="bottom">Driver</td><td align="left" valign="bottom">0.038</td><td align="left" valign="bottom">17</td><td align="left" valign="bottom">0.0028 ± 2e-05</td><td align="left" valign="bottom">50.2185 ± 0.2399</td><td align="left" valign="bottom">57.3764 ± 0.0934</td></tr><tr><td align="left" valign="bottom">19Ta</td><td align="left" valign="bottom">exponential pseudo-Moran</td><td align="left" valign="bottom">Hybrid</td><td align="left" valign="bottom">0.135</td><td align="left" valign="bottom">59</td><td align="left" valign="bottom">0.0022 ± 3e-05</td><td align="left" valign="bottom">48.3823 ± 0.242</td><td align="left" valign="bottom">57.5368 ± 0.0935</td></tr><tr><td align="left" valign="bottom">19Ta</td><td align="left" valign="bottom">exponential pseudo-Moran</td><td align="left" valign="bottom">Neutral</td><td align="left" valign="bottom">0.116</td><td align="left" valign="bottom">1</td><td align="left" valign="bottom">0.0022 ± 9e-05</td><td align="left" valign="bottom">0 ± 0</td><td align="left" valign="bottom">56.5955 ± 0.6549</td></tr><tr><td align="left" valign="bottom">19Ta</td><td align="left" valign="bottom">constant Wright-Fisher</td><td align="left" valign="bottom">Abundance</td><td align="left" valign="bottom">0.97</td><td align="left" valign="bottom">11760</td><td align="left" valign="bottom">0.0075 ± 2e-05</td><td align="left" valign="bottom">69.3863 ± 0.1735</td><td align="left" valign="bottom">59.956 ± 0.0938</td></tr><tr><td align="left" valign="bottom">19Ta</td><td align="left" valign="bottom">constant Wright-Fisher</td><td align="left" valign="bottom">Driver</td><td align="left" valign="bottom">0</td><td align="left" valign="bottom">0</td><td align="left" valign="bottom">0.0028 ± 2e-05</td><td align="left" valign="bottom">48.8413 ± 0.2392</td><td align="left" valign="bottom">56.4529 ± 0.0917</td></tr><tr><td align="left" valign="bottom">19Ta</td><td align="left" valign="bottom">constant Wright-Fisher</td><td align="left" valign="bottom">Hybrid</td><td align="left" valign="bottom">0.026</td><td align="left" valign="bottom">315</td><td align="left" valign="bottom">0.0023 ± 1e-05</td><td align="left" valign="bottom">50.8588 ± 0.2383</td><td align="left" valign="bottom">57.1031 ± 0.0925</td></tr><tr><td align="left" valign="bottom">19Ta</td><td align="left" valign="bottom">constant Wright-Fisher</td><td align="left" valign="bottom">Neutral</td><td align="left" valign="bottom">0.004</td><td align="left" valign="bottom">1</td><td align="left" valign="bottom">0.0023 ± 1e-04</td><td align="left" valign="bottom">0 ± 0</td><td align="left" valign="bottom">57.9522 ± 0.6869</td></tr><tr><td align="left" valign="bottom">19Tb</td><td align="left" valign="bottom">exponential pseudo-Moran</td><td align="left" valign="bottom">Abundance</td><td align="left" valign="bottom">0.727</td><td align="left" valign="bottom">320</td><td align="left" valign="bottom">0.0036 ± 1e-05</td><td align="left" valign="bottom">60.5885 ± 0.2085</td><td align="left" valign="bottom">59.829 ± 0.0938</td></tr><tr><td align="left" valign="bottom">19Tb</td><td align="left" valign="bottom">exponential pseudo-Moran</td><td align="left" valign="bottom">Driver</td><td align="left" valign="bottom">0.03</td><td align="left" valign="bottom">13</td><td align="left" valign="bottom">0.001 ± 1e-05</td><td align="left" valign="bottom">49.6622 ± 0.2389</td><td align="left" valign="bottom">58.6929 ± 0.0944</td></tr><tr><td align="left" valign="bottom">19Tb</td><td align="left" valign="bottom">exponential pseudo-Moran</td><td align="left" valign="bottom">Hybrid</td><td align="left" valign="bottom">0.127</td><td align="left" valign="bottom">56</td><td align="left" valign="bottom">8e-04 ± 1e-05</td><td align="left" valign="bottom">48.5237 ± 0.2322</td><td align="left" valign="bottom">58.9663 ± 0.0931</td></tr><tr><td align="left" valign="bottom">19Tb</td><td align="left" valign="bottom">exponential pseudo-Moran</td><td align="left" valign="bottom">Neutral</td><td align="left" valign="bottom">0.116</td><td align="left" valign="bottom">1</td><td align="left" valign="bottom">9e-04 ± 5e-05</td><td align="left" valign="bottom">0 ± 0</td><td align="left" valign="bottom">57.7102 ± 0.67</td></tr><tr><td align="left" valign="bottom">19Tb</td><td align="left" valign="bottom">constant Wright-Fisher</td><td align="left" valign="bottom">Abundance</td><td align="left" valign="bottom">0.979</td><td align="left" valign="bottom">47320</td><td align="left" valign="bottom">0.0068 ± 2e-05</td><td align="left" valign="bottom">69.5697 ± 0.173</td><td align="left" valign="bottom">59.9232 ± 0.0935</td></tr><tr><td align="left" valign="bottom">19Tb</td><td align="left" valign="bottom">constant Wright-Fisher</td><td align="left" valign="bottom">Driver</td><td align="left" valign="bottom">0</td><td align="left" valign="bottom">0</td><td align="left" valign="bottom">0.0012 ± 1e-05</td><td align="left" valign="bottom">48.2786 ± 0.2383</td><td align="left" valign="bottom">57.5433 ± 0.0934</td></tr><tr><td align="left" valign="bottom">19Tb</td><td align="left" valign="bottom">constant Wright-Fisher</td><td align="left" valign="bottom">Hybrid</td><td align="left" valign="bottom">0.02</td><td align="left" valign="bottom">982</td><td align="left" valign="bottom">9e-04 ± 1e-05</td><td align="left" valign="bottom">50.8162 ± 0.2357</td><td align="left" valign="bottom">58.2495 ± 0.0941</td></tr><tr><td align="left" valign="bottom">19Tb</td><td align="left" valign="bottom">constant Wright-Fisher</td><td align="left" valign="bottom">Neutral</td><td align="left" valign="bottom">0.001</td><td align="left" valign="bottom">1</td><td align="left" valign="bottom">9e-04 ± 5e-05</td><td align="left" valign="bottom">0 ± 0</td><td align="left" valign="bottom">58.8376 ± 0.669</td></tr><tr><td align="left" valign="bottom">24Ta</td><td align="left" valign="bottom">exponential pseudo-Moran</td><td align="left" valign="bottom">Abundance</td><td align="left" valign="bottom">0.731</td><td align="left" valign="bottom">321</td><td align="left" valign="bottom">0.0036 ± 1e-05</td><td align="left" valign="bottom">60.5303 ± 0.2082</td><td align="left" valign="bottom">59.8208 ± 0.0938</td></tr><tr><td align="left" valign="bottom">24Ta</td><td align="left" valign="bottom">exponential pseudo-Moran</td><td align="left" valign="bottom">Driver</td><td align="left" valign="bottom">0.029</td><td align="left" valign="bottom">13</td><td align="left" valign="bottom">0.001 ± 1e-05</td><td align="left" valign="bottom">49.6703 ± 0.2389</td><td align="left" valign="bottom">58.6938 ± 0.0944</td></tr><tr><td align="left" valign="bottom">24Ta</td><td align="left" valign="bottom">exponential pseudo-Moran</td><td align="left" valign="bottom">Hybrid</td><td align="left" valign="bottom">0.125</td><td align="left" valign="bottom">55</td><td align="left" valign="bottom">8e-04 ± 1e-05</td><td align="left" valign="bottom">49.3669 ± 0.2376</td><td align="left" valign="bottom">58.5778 ± 0.0935</td></tr><tr><td align="left" valign="bottom">24Ta</td><td align="left" valign="bottom">exponential pseudo-Moran</td><td align="left" valign="bottom">Neutral</td><td align="left" valign="bottom">0.116</td><td align="left" valign="bottom">1</td><td align="left" valign="bottom">9e-04 ± 5e-05</td><td align="left" valign="bottom">0 ± 0</td><td align="left" valign="bottom">57.7102 ± 0.67</td></tr><tr><td align="left" valign="bottom">24Ta</td><td align="left" valign="bottom">constant Wright-Fisher</td><td align="left" valign="bottom">Abundance</td><td align="left" valign="bottom">0.979</td><td align="left" valign="bottom">47346</td><td align="left" valign="bottom">0.0068 ± 2e-05</td><td align="left" valign="bottom">69.6173 ± 0.173</td><td align="left" valign="bottom">59.933 ± 0.0934</td></tr><tr><td align="left" valign="bottom">24Ta</td><td align="left" valign="bottom">constant Wright-Fisher</td><td align="left" valign="bottom">Driver</td><td align="left" valign="bottom">0</td><td align="left" valign="bottom">0</td><td align="left" valign="bottom">0.0012 ± 1e-05</td><td align="left" valign="bottom">48.2789 ± 0.2383</td><td align="left" valign="bottom">57.5377 ± 0.0934</td></tr><tr><td align="left" valign="bottom">24Ta</td><td align="left" valign="bottom">constant Wright-Fisher</td><td align="left" valign="bottom">Hybrid</td><td align="left" valign="bottom">0.02</td><td align="left" valign="bottom">956</td><td align="left" valign="bottom">9e-04 ± 1e-05</td><td align="left" valign="bottom">50.8229 ± 0.2357</td><td align="left" valign="bottom">58.2524 ± 0.0941</td></tr><tr><td align="left" valign="bottom">24Ta</td><td align="left" valign="bottom">constant Wright-Fisher</td><td align="left" valign="bottom">Neutral</td><td align="left" valign="bottom">0.001</td><td align="left" valign="bottom">1</td><td align="left" valign="bottom">9e-04 ± 5e-05</td><td align="left" valign="bottom">0 ± 0</td><td align="left" valign="bottom">58.8567 ± 0.6676</td></tr><tr><td align="left" valign="bottom">24Tb</td><td align="left" valign="bottom">exponential pseudo-Moran</td><td align="left" valign="bottom">Abundance</td><td align="left" valign="bottom">0.68</td><td align="left" valign="bottom">294</td><td align="left" valign="bottom">0.0046 ± 1e-05</td><td align="left" valign="bottom">60.2602 ± 0.2084</td><td align="left" valign="bottom">59.8073 ± 0.0936</td></tr><tr><td align="left" valign="bottom">24Tb</td><td align="left" valign="bottom">exponential pseudo-Moran</td><td align="left" valign="bottom">Driver</td><td align="left" valign="bottom">0.054</td><td align="left" valign="bottom">23</td><td align="left" valign="bottom">0.0031 ± 3e-05</td><td align="left" valign="bottom">50.2981 ± 0.2399</td><td align="left" valign="bottom">57.2927 ± 0.0934</td></tr><tr><td align="left" valign="bottom">24Tb</td><td align="left" valign="bottom">exponential pseudo-Moran</td><td align="left" valign="bottom">Hybrid</td><td align="left" valign="bottom">0.149</td><td align="left" valign="bottom">65</td><td align="left" valign="bottom">0.0025 ± 4e-05</td><td align="left" valign="bottom">48.3833 ± 0.244</td><td align="left" valign="bottom">57.4236 ± 0.0936</td></tr><tr><td align="left" valign="bottom">24Tb</td><td align="left" valign="bottom">exponential pseudo-Moran</td><td align="left" valign="bottom">Neutral</td><td align="left" valign="bottom">0.118</td><td align="left" valign="bottom">1</td><td align="left" valign="bottom">0.0025 ± 0.00013</td><td align="left" valign="bottom">0 ± 0</td><td align="left" valign="bottom">56.7229 ± 0.6579</td></tr><tr><td align="left" valign="bottom">24Tb</td><td align="left" valign="bottom">constant Wright-Fisher</td><td align="left" valign="bottom">Abundance</td><td align="left" valign="bottom">0.954</td><td align="left" valign="bottom">7730</td><td align="left" valign="bottom">0.0215 ± 0.00011</td><td align="left" valign="bottom">33.6703 ± 0.2962</td><td align="left" valign="bottom">59.9064 ± 0.0937</td></tr><tr><td align="left" valign="bottom">24Tb</td><td align="left" valign="bottom">constant Wright-Fisher</td><td align="left" valign="bottom">Driver</td><td align="left" valign="bottom">0</td><td align="left" valign="bottom">2</td><td align="left" valign="bottom">0.003 ± 2e-05</td><td align="left" valign="bottom">48.7528 ± 0.2393</td><td align="left" valign="bottom">56.4175 ± 0.0918</td></tr><tr><td align="left" valign="bottom">24Tb</td><td align="left" valign="bottom">constant Wright-Fisher</td><td align="left" valign="bottom">Hybrid</td><td align="left" valign="bottom">0.039</td><td align="left" valign="bottom">318</td><td align="left" valign="bottom">0.0024 ± 2e-05</td><td align="left" valign="bottom">50.7006 ± 0.2389</td><td align="left" valign="bottom">57.107 ± 0.0925</td></tr><tr><td align="left" valign="bottom">24Tb</td><td align="left" valign="bottom">constant Wright-Fisher</td><td align="left" valign="bottom">Neutral</td><td align="left" valign="bottom">0.006</td><td align="left" valign="bottom">1</td><td align="left" valign="bottom">0.0024 ± 0.00011</td><td align="left" valign="bottom">0 ± 0</td><td align="left" valign="bottom">58.0318 ± 0.6822</td></tr><tr><td align="left" valign="bottom">26N</td><td align="left" valign="bottom">exponential pseudo-Moran</td><td align="left" valign="bottom">Abundance</td><td align="left" valign="bottom">0.582</td><td align="left" valign="bottom">199</td><td align="left" valign="bottom">0.0021 ± 1e-05</td><td align="left" valign="bottom">60.9877 ± 0.2031</td><td align="left" valign="bottom">59.9205 ± 0.0934</td></tr><tr><td align="left" valign="bottom">26N</td><td align="left" valign="bottom">exponential pseudo-Moran</td><td align="left" valign="bottom">Driver</td><td align="left" valign="bottom">0.113</td><td align="left" valign="bottom">39</td><td align="left" valign="bottom">0.001 ± 1e-05</td><td align="left" valign="bottom">49.6389 ± 0.2389</td><td align="left" valign="bottom">58.7018 ± 0.0944</td></tr><tr><td align="left" valign="bottom">26N</td><td align="left" valign="bottom">exponential pseudo-Moran</td><td align="left" valign="bottom">Hybrid</td><td align="left" valign="bottom">0.156</td><td align="left" valign="bottom">54</td><td align="left" valign="bottom">8e-04 ± 1e-05</td><td align="left" valign="bottom">49.3389 ± 0.2377</td><td align="left" valign="bottom">58.5755 ± 0.0935</td></tr><tr><td align="left" valign="bottom">26N</td><td align="left" valign="bottom">exponential pseudo-Moran</td><td align="left" valign="bottom">Neutral</td><td align="left" valign="bottom">0.149</td><td align="left" valign="bottom">1</td><td align="left" valign="bottom">9e-04 ± 5e-05</td><td align="left" valign="bottom">0 ± 0</td><td align="left" valign="bottom">57.7994 ± 0.6728</td></tr><tr><td align="left" valign="bottom">26N</td><td align="left" valign="bottom">constant Wright-Fisher</td><td align="left" valign="bottom">Abundance</td><td align="left" valign="bottom">0.654</td><td align="left" valign="bottom">290</td><td align="left" valign="bottom">0.0039 ± 1e-05</td><td align="left" valign="bottom">69.794 ± 0.1704</td><td align="left" valign="bottom">59.9547 ± 0.0935</td></tr><tr><td align="left" valign="bottom">26N</td><td align="left" valign="bottom">constant Wright-Fisher</td><td align="left" valign="bottom">Driver</td><td align="left" valign="bottom">0.115</td><td align="left" valign="bottom">51</td><td align="left" valign="bottom">0.0012 ± 1e-05</td><td align="left" valign="bottom">48.2849 ± 0.2384</td><td align="left" valign="bottom">57.5175 ± 0.0933</td></tr><tr><td align="left" valign="bottom">26N</td><td align="left" valign="bottom">constant Wright-Fisher</td><td align="left" valign="bottom">Hybrid</td><td align="left" valign="bottom">0.115</td><td align="left" valign="bottom">51</td><td align="left" valign="bottom">9e-04 ± 1e-05</td><td align="left" valign="bottom">50.737 ± 0.2359</td><td align="left" valign="bottom">58.2609 ± 0.0941</td></tr><tr><td align="left" valign="bottom">26N</td><td align="left" valign="bottom">constant Wright-Fisher</td><td align="left" valign="bottom">Neutral</td><td align="left" valign="bottom">0.115</td><td align="left" valign="bottom">1</td><td align="left" valign="bottom">9e-04 ± 5e-05</td><td align="left" valign="bottom">0 ± 0</td><td align="left" valign="bottom">58.7803 ± 0.6701</td></tr><tr><td align="left" valign="bottom">9T</td><td align="left" valign="bottom">exponential pseudo-Moran</td><td align="left" valign="bottom">Abundance</td><td align="left" valign="bottom">0.685</td><td align="left" valign="bottom">299</td><td align="left" valign="bottom">0.0044 ± 1e-05</td><td align="left" valign="bottom">60.2829 ± 0.2086</td><td align="left" valign="bottom">59.7955 ± 0.0936</td></tr><tr><td align="left" valign="bottom">9T</td><td align="left" valign="bottom">exponential pseudo-Moran</td><td align="left" valign="bottom">Driver</td><td align="left" valign="bottom">0.052</td><td align="left" valign="bottom">23</td><td align="left" valign="bottom">0.0029 ± 2e-05</td><td align="left" valign="bottom">50.2323 ± 0.2398</td><td align="left" valign="bottom">57.3657 ± 0.0934</td></tr><tr><td align="left" valign="bottom">9T</td><td align="left" valign="bottom">exponential pseudo-Moran</td><td align="left" valign="bottom">Hybrid</td><td align="left" valign="bottom">0.147</td><td align="left" valign="bottom">64</td><td align="left" valign="bottom">0.0022 ± 3e-05</td><td align="left" valign="bottom">48.3829 ± 0.2422</td><td align="left" valign="bottom">57.5193 ± 0.0936</td></tr><tr><td align="left" valign="bottom">9T</td><td align="left" valign="bottom">exponential pseudo-Moran</td><td align="left" valign="bottom">Neutral</td><td align="left" valign="bottom">0.117</td><td align="left" valign="bottom">1</td><td align="left" valign="bottom">0.0023 ± 9e-05</td><td align="left" valign="bottom">0 ± 0</td><td align="left" valign="bottom">56.6083 ± 0.6581</td></tr><tr><td align="left" valign="bottom">9T</td><td align="left" valign="bottom">constant Wright-Fisher</td><td align="left" valign="bottom">Abundance</td><td align="left" valign="bottom">0.958</td><td align="left" valign="bottom">9299</td><td align="left" valign="bottom">0.0087 ± 2e-05</td><td align="left" valign="bottom">69.6836 ± 0.1724</td><td align="left" valign="bottom">59.926 ± 0.0937</td></tr><tr><td align="left" valign="bottom">9T</td><td align="left" valign="bottom">constant Wright-Fisher</td><td align="left" valign="bottom">Driver</td><td align="left" valign="bottom">0</td><td align="left" valign="bottom">1</td><td align="left" valign="bottom">0.0028 ± 2e-05</td><td align="left" valign="bottom">48.8394 ± 0.2392</td><td align="left" valign="bottom">56.4465 ± 0.0917</td></tr><tr><td align="left" valign="bottom">9T</td><td align="left" valign="bottom">constant Wright-Fisher</td><td align="left" valign="bottom">Hybrid</td><td align="left" valign="bottom">0.037</td><td align="left" valign="bottom">360</td><td align="left" valign="bottom">0.0023 ± 1e-05</td><td align="left" valign="bottom">50.8477 ± 0.2384</td><td align="left" valign="bottom">57.0952 ± 0.0925</td></tr><tr><td align="left" valign="bottom">9T</td><td align="left" valign="bottom">constant Wright-Fisher</td><td align="left" valign="bottom">Neutral</td><td align="left" valign="bottom">0.005</td><td align="left" valign="bottom">1</td><td align="left" valign="bottom">0.0023 ± 1e-04</td><td align="left" valign="bottom">0 ± 0</td><td align="left" valign="bottom">57.9427 ± 0.687</td></tr><tr><td align="left" valign="bottom">PolyB1</td><td align="left" valign="bottom">exponential pseudo-Moran</td><td align="left" valign="bottom">Abundance</td><td align="left" valign="bottom">0.635</td><td align="left" valign="bottom">261</td><td align="left" valign="bottom">0.0053 ± 1e-05</td><td align="left" valign="bottom">59.5088 ± 0.2104</td><td align="left" valign="bottom">59.8379 ± 0.0935</td></tr><tr><td align="left" valign="bottom">PolyB1</td><td align="left" valign="bottom">exponential pseudo-Moran</td><td align="left" valign="bottom">Driver</td><td align="left" valign="bottom">0.076</td><td align="left" valign="bottom">31</td><td align="left" valign="bottom">0.0028 ± 2e-05</td><td align="left" valign="bottom">50.2364 ± 0.2398</td><td align="left" valign="bottom">57.4025 ± 0.0934</td></tr><tr><td align="left" valign="bottom">PolyB1</td><td align="left" valign="bottom">exponential pseudo-Moran</td><td align="left" valign="bottom">Hybrid</td><td align="left" valign="bottom">0.164</td><td align="left" valign="bottom">67</td><td align="left" valign="bottom">0.0022 ± 3e-05</td><td align="left" valign="bottom">48.6949 ± 0.2419</td><td align="left" valign="bottom">57.6322 ± 0.0934</td></tr><tr><td align="left" valign="bottom">PolyB1</td><td align="left" valign="bottom">exponential pseudo-Moran</td><td align="left" valign="bottom">Neutral</td><td align="left" valign="bottom">0.124</td><td align="left" valign="bottom">1</td><td align="left" valign="bottom">0.0022 ± 9e-05</td><td align="left" valign="bottom">0 ± 0</td><td align="left" valign="bottom">56.5955 ± 0.6549</td></tr><tr><td align="left" valign="bottom">PolyB1</td><td align="left" valign="bottom">constant Wright-Fisher</td><td align="left" valign="bottom">Abundance</td><td align="left" valign="bottom">0.925</td><td align="left" valign="bottom">3482</td><td align="left" valign="bottom">0.0111 ± 3e-05</td><td align="left" valign="bottom">70.2557 ± 0.169</td><td align="left" valign="bottom">60.042 ± 0.0936</td></tr><tr><td align="left" valign="bottom">PolyB1</td><td align="left" valign="bottom">constant Wright-Fisher</td><td align="left" valign="bottom">Driver</td><td align="left" valign="bottom">0.001</td><td align="left" valign="bottom">4</td><td align="left" valign="bottom">0.0028 ± 2e-05</td><td align="left" valign="bottom">48.8194 ± 0.2391</td><td align="left" valign="bottom">56.4451 ± 0.0917</td></tr><tr><td align="left" valign="bottom">PolyB1</td><td align="left" valign="bottom">constant Wright-Fisher</td><td align="left" valign="bottom">Hybrid</td><td align="left" valign="bottom">0.061</td><td align="left" valign="bottom">228</td><td align="left" valign="bottom">0.0023 ± 1e-05</td><td align="left" valign="bottom">50.895 ± 0.2381</td><td align="left" valign="bottom">57.1073 ± 0.0925</td></tr><tr><td align="left" valign="bottom">PolyB1</td><td align="left" valign="bottom">constant Wright-Fisher</td><td align="left" valign="bottom">Neutral</td><td align="left" valign="bottom">0.014</td><td align="left" valign="bottom">1</td><td align="left" valign="bottom">0.0023 ± 1e-04</td><td align="left" valign="bottom">0 ± 0</td><td align="left" valign="bottom">57.9809 ± 0.6861</td></tr><tr><td align="left" valign="bottom">PolyB2</td><td align="left" valign="bottom">exponential pseudo-Moran</td><td align="left" valign="bottom">Abundance</td><td align="left" valign="bottom">0.603</td><td align="left" valign="bottom">218</td><td align="left" valign="bottom">0.0059 ± 1e-05</td><td align="left" valign="bottom">58.6612 ± 0.212</td><td align="left" valign="bottom">59.7835 ± 0.0937</td></tr><tr><td align="left" valign="bottom">PolyB2</td><td align="left" valign="bottom">exponential pseudo-Moran</td><td align="left" valign="bottom">Driver</td><td align="left" valign="bottom">0.086</td><td align="left" valign="bottom">31</td><td align="left" valign="bottom">0.0038 ± 4e-05</td><td align="left" valign="bottom">50.2948 ± 0.2394</td><td align="left" valign="bottom">57.0217 ± 0.093</td></tr><tr><td align="left" valign="bottom">PolyB2</td><td align="left" valign="bottom">exponential pseudo-Moran</td><td align="left" valign="bottom">Hybrid</td><td align="left" valign="bottom">0.17</td><td align="left" valign="bottom">61</td><td align="left" valign="bottom">0.004 ± 7e-05</td><td align="left" valign="bottom">48.9466 ± 0.2472</td><td align="left" valign="bottom">57.28 ± 0.0942</td></tr><tr><td align="left" valign="bottom">PolyB2</td><td align="left" valign="bottom">exponential pseudo-Moran</td><td align="left" valign="bottom">Neutral</td><td align="left" valign="bottom">0.141</td><td align="left" valign="bottom">1</td><td align="left" valign="bottom">0.0033 ± 0.00022</td><td align="left" valign="bottom">0 ± 0</td><td align="left" valign="bottom">56.5732 ± 0.6597</td></tr><tr><td align="left" valign="bottom">PolyB2</td><td align="left" valign="bottom">constant Wright-Fisher</td><td align="left" valign="bottom">Abundance</td><td align="left" valign="bottom">0.893</td><td align="left" valign="bottom">1277</td><td align="left" valign="bottom">0.0301 ± 1e-04</td><td align="left" valign="bottom">3.0543 ± 0.0165</td><td align="left" valign="bottom">59.9142 ± 0.0936</td></tr><tr><td align="left" valign="bottom">PolyB2</td><td align="left" valign="bottom">constant Wright-Fisher</td><td align="left" valign="bottom">Driver</td><td align="left" valign="bottom">0.003</td><td align="left" valign="bottom">4</td><td align="left" valign="bottom">0.0034 ± 3e-05</td><td align="left" valign="bottom">48.7328 ± 0.2396</td><td align="left" valign="bottom">56.3664 ± 0.0917</td></tr><tr><td align="left" valign="bottom">PolyB2</td><td align="left" valign="bottom">constant Wright-Fisher</td><td align="left" valign="bottom">Hybrid</td><td align="left" valign="bottom">0.069</td><td align="left" valign="bottom">98</td><td align="left" valign="bottom">0.0027 ± 2e-05</td><td align="left" valign="bottom">50.3534 ± 0.2405</td><td align="left" valign="bottom">57.1445 ± 0.0928</td></tr><tr><td align="left" valign="bottom">PolyB2</td><td align="left" valign="bottom">constant Wright-Fisher</td><td align="left" valign="bottom">Neutral</td><td align="left" valign="bottom">0.036</td><td align="left" valign="bottom">1</td><td align="left" valign="bottom">0.0026 ± 0.00014</td><td align="left" valign="bottom">0 ± 0</td><td align="left" valign="bottom">58.1592 ± 0.6741</td></tr></tbody></table></table-wrap><p>Having confirmed the summary statistics from these samples were within the space of the simulation data with our chosen priors (<xref ref-type="fig" rid="fig6">Figure 6B</xref>), we performed ABC analysis on these datasets to infer rates of chromosome mis-segregation and levels of selection pressure and display the joint posterior distributions as 2D density plots (<xref ref-type="fig" rid="fig6">Figure 6C and D</xref>; <xref ref-type="fig" rid="fig6s2">Figure 6—figure supplements 2</xref> and <xref ref-type="fig" rid="fig6s3">3</xref>). <xref ref-type="fig" rid="fig6">Figure 6C</xref> illustrates the results for two individual colon organoid lines, showing the distribution of parameters used for simulations that gave the most similar results. With ABC, inferred parameters fall within rates of mis-segregation of about 0.001–0.006. Applied to a near-diploid cell, this translates to a range of about 5–38% of cell divisions having one chromosome mis-segregation. Importantly, these inferred rates of chromosome mis-segregation fall within the range of approximated <italic>per chromosome</italic> rates experimentally observed in cancer cell lines and human tumors (<xref ref-type="fig" rid="fig6">Figure 6E</xref>;<xref ref-type="table" rid="table4">Table 4</xref>, <xref ref-type="table" rid="table5">Table 5</xref>; <xref ref-type="bibr" rid="bib4">Bakhoum et al., 2014</xref>; <xref ref-type="bibr" rid="bib3">Bakhoum et al., 2011</xref>; <xref ref-type="bibr" rid="bib2">Bakhoum et al., 2009</xref>; <xref ref-type="bibr" rid="bib19">Dewhurst et al., 2014</xref>; <xref ref-type="bibr" rid="bib55">Nicholson et al., 2015</xref>; <xref ref-type="bibr" rid="bib57">Orr et al., 2016</xref>; <xref ref-type="bibr" rid="bib74">Thompson and Compton, 2008</xref>; <xref ref-type="bibr" rid="bib82">Worrall et al., 2018</xref>; <xref ref-type="bibr" rid="bib83">Zasadil et al., 2014</xref>). Higher inferred mis-segregation rates tended to coincide with lower inferred selection experienced in these samples (<xref ref-type="fig" rid="fig6">Figure 6F</xref>). Posterior distributions in these samples were skewed toward high selection (S) indicating the presence stabilizing selection in all cases, where the average of the distributions of some samples were slightly lower or higher (<xref ref-type="fig" rid="fig6s3">Figure 6—figure supplement 3</xref>).</p><table-wrap id="table4" position="float"><label>Table 4.</label><caption><title>Model selection with selective pressure constrained to S = 1.</title></caption><table frame="hsides" rules="groups"><thead><tr><th align="left" valign="bottom">Sample</th><th align="left" valign="bottom">Growth Model</th><th align="left" valign="bottom">Selection Model</th><th align="left" valign="bottom">PP</th><th align="left" valign="bottom">BF (Ho Neutral)</th><th align="left" valign="bottom">Pmisseg</th><th align="left" valign="bottom">S</th><th align="left" valign="bottom">Steps</th></tr></thead><tbody><tr><td align="left" valign="bottom">7T</td><td align="left" valign="bottom">exponential pseudo-Moran</td><td align="left" valign="bottom">Abundance</td><td align="left" valign="bottom">0.274</td><td align="left" valign="bottom">1</td><td align="left" valign="bottom">9e-04 ± 5e-05</td><td align="left" valign="bottom">1 ± 0</td><td align="left" valign="bottom">58.2452 ± 0.6646</td></tr><tr><td align="left" valign="bottom">7T</td><td align="left" valign="bottom">exponential pseudo-Moran</td><td align="left" valign="bottom">Driver</td><td align="left" valign="bottom">0.238</td><td align="left" valign="bottom">1</td><td align="left" valign="bottom">9e-04 ± 5e-05</td><td align="left" valign="bottom">1 ± 0</td><td align="left" valign="bottom">58.4745 ± 0.6725</td></tr><tr><td align="left" valign="bottom">7T</td><td align="left" valign="bottom">exponential pseudo-Moran</td><td align="left" valign="bottom">Hybrid</td><td align="left" valign="bottom">0.26</td><td align="left" valign="bottom">1</td><td align="left" valign="bottom">9e-04 ± 5e-05</td><td align="left" valign="bottom">1 ± 0</td><td align="left" valign="bottom">58.586 ± 0.6668</td></tr><tr><td align="left" valign="bottom">7T</td><td align="left" valign="bottom">exponential pseudo-Moran</td><td align="left" valign="bottom">Neutral</td><td align="left" valign="bottom">0.228</td><td align="left" valign="bottom">1</td><td align="left" valign="bottom">9e-04 ± 6e-05</td><td align="left" valign="bottom">1 ± 0</td><td align="left" valign="bottom">58.5446 ± 0.6791</td></tr><tr><td align="left" valign="bottom">7T</td><td align="left" valign="bottom">constant Wright-Fisher</td><td align="left" valign="bottom">Abundance</td><td align="left" valign="bottom">0.259</td><td align="left" valign="bottom">1</td><td align="left" valign="bottom">9e-04 ± 6e-05</td><td align="left" valign="bottom">1 ± 0</td><td align="left" valign="bottom">58.8089 ± 0.6627</td></tr><tr><td align="left" valign="bottom">7T</td><td align="left" valign="bottom">constant Wright-Fisher</td><td align="left" valign="bottom">Driver</td><td align="left" valign="bottom">0.24</td><td align="left" valign="bottom">1</td><td align="left" valign="bottom">9e-04 ± 6e-05</td><td align="left" valign="bottom">1 ± 0</td><td align="left" valign="bottom">58.1783 ± 0.6771</td></tr><tr><td align="left" valign="bottom">7T</td><td align="left" valign="bottom">constant Wright-Fisher</td><td align="left" valign="bottom">Hybrid</td><td align="left" valign="bottom">0.257</td><td align="left" valign="bottom">1</td><td align="left" valign="bottom">9e-04 ± 5e-05</td><td align="left" valign="bottom">1 ± 0</td><td align="left" valign="bottom">59.0924 ± 0.6742</td></tr><tr><td align="left" valign="bottom">7T</td><td align="left" valign="bottom">constant Wright-Fisher</td><td align="left" valign="bottom">Neutral</td><td align="left" valign="bottom">0.245</td><td align="left" valign="bottom">1</td><td align="left" valign="bottom">9e-04 ± 7e-05</td><td align="left" valign="bottom">1 ± 0</td><td align="left" valign="bottom">58.7516 ± 0.6787</td></tr><tr><td align="left" valign="bottom">U1T</td><td align="left" valign="bottom">exponential pseudo-Moran</td><td align="left" valign="bottom">Abundance</td><td align="left" valign="bottom">0.275</td><td align="left" valign="bottom">1</td><td align="left" valign="bottom">9e-04 ± 5e-05</td><td align="left" valign="bottom">1 ± 0</td><td align="left" valign="bottom">58.2452 ± 0.6646</td></tr><tr><td align="left" valign="bottom">U1T</td><td align="left" valign="bottom">exponential pseudo-Moran</td><td align="left" valign="bottom">Driver</td><td align="left" valign="bottom">0.239</td><td align="left" valign="bottom">1</td><td align="left" valign="bottom">9e-04 ± 5e-05</td><td align="left" valign="bottom">1 ± 0</td><td align="left" valign="bottom">58.4745 ± 0.6725</td></tr><tr><td align="left" valign="bottom">U1T</td><td align="left" valign="bottom">exponential pseudo-Moran</td><td align="left" valign="bottom">Hybrid</td><td align="left" valign="bottom">0.258</td><td align="left" valign="bottom">1</td><td align="left" valign="bottom">9e-04 ± 5e-05</td><td align="left" valign="bottom">1 ± 0</td><td align="left" valign="bottom">58.586 ± 0.6668</td></tr><tr><td align="left" valign="bottom">U1T</td><td align="left" valign="bottom">exponential pseudo-Moran</td><td align="left" valign="bottom">Neutral</td><td align="left" valign="bottom">0.228</td><td align="left" valign="bottom">1</td><td align="left" valign="bottom">9e-04 ± 6e-05</td><td align="left" valign="bottom">1 ± 0</td><td align="left" valign="bottom">58.5446 ± 0.6791</td></tr><tr><td align="left" valign="bottom">U1T</td><td align="left" valign="bottom">constant Wright-Fisher</td><td align="left" valign="bottom">Abundance</td><td align="left" valign="bottom">0.259</td><td align="left" valign="bottom">1</td><td align="left" valign="bottom">9e-04 ± 6e-05</td><td align="left" valign="bottom">1 ± 0</td><td align="left" valign="bottom">58.8089 ± 0.6627</td></tr><tr><td align="left" valign="bottom">U1T</td><td align="left" valign="bottom">constant Wright-Fisher</td><td align="left" valign="bottom">Driver</td><td align="left" valign="bottom">0.24</td><td align="left" valign="bottom">1</td><td align="left" valign="bottom">9e-04 ± 6e-05</td><td align="left" valign="bottom">1 ± 0</td><td align="left" valign="bottom">58.1783 ± 0.6771</td></tr><tr><td align="left" valign="bottom">U1T</td><td align="left" valign="bottom">constant Wright-Fisher</td><td align="left" valign="bottom">Hybrid</td><td align="left" valign="bottom">0.257</td><td align="left" valign="bottom">1</td><td align="left" valign="bottom">9e-04 ± 5e-05</td><td align="left" valign="bottom">1 ± 0</td><td align="left" valign="bottom">59.1592 ± 0.6715</td></tr><tr><td align="left" valign="bottom">U1T</td><td align="left" valign="bottom">constant Wright-Fisher</td><td align="left" valign="bottom">Neutral</td><td align="left" valign="bottom">0.245</td><td align="left" valign="bottom">1</td><td align="left" valign="bottom">9e-04 ± 7e-05</td><td align="left" valign="bottom">1 ± 0</td><td align="left" valign="bottom">58.7516 ± 0.6787</td></tr><tr><td align="left" valign="bottom">U2T</td><td align="left" valign="bottom">exponential pseudo-Moran</td><td align="left" valign="bottom">Abundance</td><td align="left" valign="bottom">0.276</td><td align="left" valign="bottom">1</td><td align="left" valign="bottom">0.0021 ± 8e-05</td><td align="left" valign="bottom">1 ± 0</td><td align="left" valign="bottom">57.3057 ± 0.653</td></tr><tr><td align="left" valign="bottom">U2T</td><td align="left" valign="bottom">exponential pseudo-Moran</td><td align="left" valign="bottom">Driver</td><td align="left" valign="bottom">0.235</td><td align="left" valign="bottom">1</td><td align="left" valign="bottom">0.0024 ± 0.00011</td><td align="left" valign="bottom">1 ± 0</td><td align="left" valign="bottom">57.7452 ± 0.6634</td></tr><tr><td align="left" valign="bottom">U2T</td><td align="left" valign="bottom">exponential pseudo-Moran</td><td align="left" valign="bottom">Hybrid</td><td align="left" valign="bottom">0.264</td><td align="left" valign="bottom">1</td><td align="left" valign="bottom">0.0021 ± 7e-05</td><td align="left" valign="bottom">1 ± 0</td><td align="left" valign="bottom">58.1274 ± 0.654</td></tr><tr><td align="left" valign="bottom">U2T</td><td align="left" valign="bottom">exponential pseudo-Moran</td><td align="left" valign="bottom">Neutral</td><td align="left" valign="bottom">0.225</td><td align="left" valign="bottom">1</td><td align="left" valign="bottom">0.0024 ± 0.00011</td><td align="left" valign="bottom">1 ± 0</td><td align="left" valign="bottom">57.8758 ± 0.6772</td></tr><tr><td align="left" valign="bottom">U2T</td><td align="left" valign="bottom">constant Wright-Fisher</td><td align="left" valign="bottom">Abundance</td><td align="left" valign="bottom">0.269</td><td align="left" valign="bottom">1</td><td align="left" valign="bottom">0.0023 ± 1e-04</td><td align="left" valign="bottom">1 ± 0</td><td align="left" valign="bottom">58.3439 ± 0.6532</td></tr><tr><td align="left" valign="bottom">U2T</td><td align="left" valign="bottom">constant Wright-Fisher</td><td align="left" valign="bottom">Driver</td><td align="left" valign="bottom">0.233</td><td align="left" valign="bottom">1</td><td align="left" valign="bottom">0.0023 ± 9e-05</td><td align="left" valign="bottom">1 ± 0</td><td align="left" valign="bottom">57.4777 ± 0.693</td></tr><tr><td align="left" valign="bottom">U2T</td><td align="left" valign="bottom">constant Wright-Fisher</td><td align="left" valign="bottom">Hybrid</td><td align="left" valign="bottom">0.263</td><td align="left" valign="bottom">1</td><td align="left" valign="bottom">0.0023 ± 1e-04</td><td align="left" valign="bottom">1 ± 0</td><td align="left" valign="bottom">57.8662 ± 0.6683</td></tr><tr><td align="left" valign="bottom">U2T</td><td align="left" valign="bottom">constant Wright-Fisher</td><td align="left" valign="bottom">Neutral</td><td align="left" valign="bottom">0.236</td><td align="left" valign="bottom">1</td><td align="left" valign="bottom">0.0025 ± 0.00012</td><td align="left" valign="bottom">1 ± 0</td><td align="left" valign="bottom">57.1433 ± 0.6655</td></tr><tr><td align="left" valign="bottom">U3T</td><td align="left" valign="bottom">exponential pseudo-Moran</td><td align="left" valign="bottom">Abundance</td><td align="left" valign="bottom">0.275</td><td align="left" valign="bottom">1</td><td align="left" valign="bottom">9e-04 ± 5e-05</td><td align="left" valign="bottom">1 ± 0</td><td align="left" valign="bottom">58.1624 ± 0.6643</td></tr><tr><td align="left" valign="bottom">U3T</td><td align="left" valign="bottom">exponential pseudo-Moran</td><td align="left" valign="bottom">Driver</td><td align="left" valign="bottom">0.239</td><td align="left" valign="bottom">1</td><td align="left" valign="bottom">9e-04 ± 5e-05</td><td align="left" valign="bottom">1 ± 0</td><td align="left" valign="bottom">58.4554 ± 0.6736</td></tr><tr><td align="left" valign="bottom">U3T</td><td align="left" valign="bottom">exponential pseudo-Moran</td><td align="left" valign="bottom">Hybrid</td><td align="left" valign="bottom">0.258</td><td align="left" valign="bottom">1</td><td align="left" valign="bottom">9e-04 ± 5e-05</td><td align="left" valign="bottom">1 ± 0</td><td align="left" valign="bottom">58.586 ± 0.6668</td></tr><tr><td align="left" valign="bottom">U3T</td><td align="left" valign="bottom">exponential pseudo-Moran</td><td align="left" valign="bottom">Neutral</td><td align="left" valign="bottom">0.228</td><td align="left" valign="bottom">1</td><td align="left" valign="bottom">9e-04 ± 6e-05</td><td align="left" valign="bottom">1 ± 0</td><td align="left" valign="bottom">58.6178 ± 0.6777</td></tr><tr><td align="left" valign="bottom">U3T</td><td align="left" valign="bottom">constant Wright-Fisher</td><td align="left" valign="bottom">Abundance</td><td align="left" valign="bottom">0.259</td><td align="left" valign="bottom">1</td><td align="left" valign="bottom">9e-04 ± 6e-05</td><td align="left" valign="bottom">1 ± 0</td><td align="left" valign="bottom">58.7611 ± 0.6614</td></tr><tr><td align="left" valign="bottom">U3T</td><td align="left" valign="bottom">constant Wright-Fisher</td><td align="left" valign="bottom">Driver</td><td align="left" valign="bottom">0.24</td><td align="left" valign="bottom">1</td><td align="left" valign="bottom">9e-04 ± 6e-05</td><td align="left" valign="bottom">1 ± 0</td><td align="left" valign="bottom">58.1783 ± 0.6771</td></tr><tr><td align="left" valign="bottom">U3T</td><td align="left" valign="bottom">constant Wright-Fisher</td><td align="left" valign="bottom">Hybrid</td><td align="left" valign="bottom">0.257</td><td align="left" valign="bottom">1</td><td align="left" valign="bottom">9e-04 ± 5e-05</td><td align="left" valign="bottom">1 ± 0</td><td align="left" valign="bottom">59.0955 ± 0.674</td></tr><tr><td align="left" valign="bottom">U3T</td><td align="left" valign="bottom">constant Wright-Fisher</td><td align="left" valign="bottom">Neutral</td><td align="left" valign="bottom">0.245</td><td align="left" valign="bottom">1</td><td align="left" valign="bottom">9e-04 ± 7e-05</td><td align="left" valign="bottom">1 ± 0</td><td align="left" valign="bottom">58.7516 ± 0.6787</td></tr><tr><td align="left" valign="bottom">14T</td><td align="left" valign="bottom">exponential pseudo-Moran</td><td align="left" valign="bottom">Abundance</td><td align="left" valign="bottom">0.275</td><td align="left" valign="bottom">1</td><td align="left" valign="bottom">9e-04 ± 5e-05</td><td align="left" valign="bottom">1 ± 0</td><td align="left" valign="bottom">58.1624 ± 0.6643</td></tr><tr><td align="left" valign="bottom">14T</td><td align="left" valign="bottom">exponential pseudo-Moran</td><td align="left" valign="bottom">Driver</td><td align="left" valign="bottom">0.239</td><td align="left" valign="bottom">1</td><td align="left" valign="bottom">9e-04 ± 5e-05</td><td align="left" valign="bottom">1 ± 0</td><td align="left" valign="bottom">58.4554 ± 0.6736</td></tr><tr><td align="left" valign="bottom">14T</td><td align="left" valign="bottom">exponential pseudo-Moran</td><td align="left" valign="bottom">Hybrid</td><td align="left" valign="bottom">0.258</td><td align="left" valign="bottom">1</td><td align="left" valign="bottom">9e-04 ± 5e-05</td><td align="left" valign="bottom">1 ± 0</td><td align="left" valign="bottom">58.586 ± 0.6668</td></tr><tr><td align="left" valign="bottom">14T</td><td align="left" valign="bottom">exponential pseudo-Moran</td><td align="left" valign="bottom">Neutral</td><td align="left" valign="bottom">0.228</td><td align="left" valign="bottom">1</td><td align="left" valign="bottom">9e-04 ± 6e-05</td><td align="left" valign="bottom">1 ± 0</td><td align="left" valign="bottom">58.5446 ± 0.6791</td></tr><tr><td align="left" valign="bottom">14T</td><td align="left" valign="bottom">constant Wright-Fisher</td><td align="left" valign="bottom">Abundance</td><td align="left" valign="bottom">0.259</td><td align="left" valign="bottom">1</td><td align="left" valign="bottom">9e-04 ± 6e-05</td><td align="left" valign="bottom">1 ± 0</td><td align="left" valign="bottom">58.8089 ± 0.6627</td></tr><tr><td align="left" valign="bottom">14T</td><td align="left" valign="bottom">constant Wright-Fisher</td><td align="left" valign="bottom">Driver</td><td align="left" valign="bottom">0.24</td><td align="left" valign="bottom">1</td><td align="left" valign="bottom">9e-04 ± 6e-05</td><td align="left" valign="bottom">1 ± 0</td><td align="left" valign="bottom">58.1783 ± 0.6771</td></tr><tr><td align="left" valign="bottom">14T</td><td align="left" valign="bottom">constant Wright-Fisher</td><td align="left" valign="bottom">Hybrid</td><td align="left" valign="bottom">0.257</td><td align="left" valign="bottom">1</td><td align="left" valign="bottom">9e-04 ± 5e-05</td><td align="left" valign="bottom">1 ± 0</td><td align="left" valign="bottom">59.0924 ± 0.6739</td></tr><tr><td align="left" valign="bottom">14T</td><td align="left" valign="bottom">constant Wright-Fisher</td><td align="left" valign="bottom">Neutral</td><td align="left" valign="bottom">0.245</td><td align="left" valign="bottom">1</td><td align="left" valign="bottom">9e-04 ± 7e-05</td><td align="left" valign="bottom">1 ± 0</td><td align="left" valign="bottom">58.7516 ± 0.6787</td></tr><tr><td align="left" valign="bottom">16T</td><td align="left" valign="bottom">exponential pseudo-Moran</td><td align="left" valign="bottom">Abundance</td><td align="left" valign="bottom">0.274</td><td align="left" valign="bottom">1</td><td align="left" valign="bottom">9e-04 ± 5e-05</td><td align="left" valign="bottom">1 ± 0</td><td align="left" valign="bottom">58.2452 ± 0.6646</td></tr><tr><td align="left" valign="bottom">16T</td><td align="left" valign="bottom">exponential pseudo-Moran</td><td align="left" valign="bottom">Driver</td><td align="left" valign="bottom">0.238</td><td align="left" valign="bottom">1</td><td align="left" valign="bottom">9e-04 ± 5e-05</td><td align="left" valign="bottom">1 ± 0</td><td align="left" valign="bottom">58.4745 ± 0.6725</td></tr><tr><td align="left" valign="bottom">16T</td><td align="left" valign="bottom">exponential pseudo-Moran</td><td align="left" valign="bottom">Hybrid</td><td align="left" valign="bottom">0.26</td><td align="left" valign="bottom">1</td><td align="left" valign="bottom">9e-04 ± 5e-05</td><td align="left" valign="bottom">1 ± 0</td><td align="left" valign="bottom">58.586 ± 0.6668</td></tr><tr><td align="left" valign="bottom">16T</td><td align="left" valign="bottom">exponential pseudo-Moran</td><td align="left" valign="bottom">Neutral</td><td align="left" valign="bottom">0.228</td><td align="left" valign="bottom">1</td><td align="left" valign="bottom">0.001 ± 6e-05</td><td align="left" valign="bottom">1 ± 0</td><td align="left" valign="bottom">58.6274 ± 0.6789</td></tr><tr><td align="left" valign="bottom">16T</td><td align="left" valign="bottom">constant Wright-Fisher</td><td align="left" valign="bottom">Abundance</td><td align="left" valign="bottom">0.259</td><td align="left" valign="bottom">1</td><td align="left" valign="bottom">9e-04 ± 6e-05</td><td align="left" valign="bottom">1 ± 0</td><td align="left" valign="bottom">58.8089 ± 0.6627</td></tr><tr><td align="left" valign="bottom">16T</td><td align="left" valign="bottom">constant Wright-Fisher</td><td align="left" valign="bottom">Driver</td><td align="left" valign="bottom">0.24</td><td align="left" valign="bottom">1</td><td align="left" valign="bottom">9e-04 ± 6e-05</td><td align="left" valign="bottom">1 ± 0</td><td align="left" valign="bottom">58.1783 ± 0.6771</td></tr><tr><td align="left" valign="bottom">16T</td><td align="left" valign="bottom">constant Wright-Fisher</td><td align="left" valign="bottom">Hybrid</td><td align="left" valign="bottom">0.257</td><td align="left" valign="bottom">1</td><td align="left" valign="bottom">9e-04 ± 5e-05</td><td align="left" valign="bottom">1 ± 0</td><td align="left" valign="bottom">59.1051 ± 0.6742</td></tr><tr><td align="left" valign="bottom">16T</td><td align="left" valign="bottom">constant Wright-Fisher</td><td align="left" valign="bottom">Neutral</td><td align="left" valign="bottom">0.245</td><td align="left" valign="bottom">1</td><td align="left" valign="bottom">9e-04 ± 7e-05</td><td align="left" valign="bottom">1 ± 0</td><td align="left" valign="bottom">58.7516 ± 0.6787</td></tr><tr><td align="left" valign="bottom">19Ta</td><td align="left" valign="bottom">exponential pseudo-Moran</td><td align="left" valign="bottom">Abundance</td><td align="left" valign="bottom">0.273</td><td align="left" valign="bottom">1</td><td align="left" valign="bottom">0.0021 ± 8e-05</td><td align="left" valign="bottom">1 ± 0</td><td align="left" valign="bottom">57.4045 ± 0.6565</td></tr><tr><td align="left" valign="bottom">19Ta</td><td align="left" valign="bottom">exponential pseudo-Moran</td><td align="left" valign="bottom">Driver</td><td align="left" valign="bottom">0.243</td><td align="left" valign="bottom">1</td><td align="left" valign="bottom">0.0024 ± 0.00011</td><td align="left" valign="bottom">1 ± 0</td><td align="left" valign="bottom">57.8025 ± 0.663</td></tr><tr><td align="left" valign="bottom">19Ta</td><td align="left" valign="bottom">exponential pseudo-Moran</td><td align="left" valign="bottom">Hybrid</td><td align="left" valign="bottom">0.261</td><td align="left" valign="bottom">1</td><td align="left" valign="bottom">0.0022 ± 8e-05</td><td align="left" valign="bottom">1 ± 0</td><td align="left" valign="bottom">57.9108 ± 0.65</td></tr><tr><td align="left" valign="bottom">19Ta</td><td align="left" valign="bottom">exponential pseudo-Moran</td><td align="left" valign="bottom">Neutral</td><td align="left" valign="bottom">0.222</td><td align="left" valign="bottom">1</td><td align="left" valign="bottom">0.0025 ± 0.00012</td><td align="left" valign="bottom">1 ± 0</td><td align="left" valign="bottom">57.9331 ± 0.6777</td></tr><tr><td align="left" valign="bottom">19Ta</td><td align="left" valign="bottom">constant Wright-Fisher</td><td align="left" valign="bottom">Abundance</td><td align="left" valign="bottom">0.27</td><td align="left" valign="bottom">1</td><td align="left" valign="bottom">0.0024 ± 0.00011</td><td align="left" valign="bottom">1 ± 0</td><td align="left" valign="bottom">58.2866 ± 0.6566</td></tr><tr><td align="left" valign="bottom">19Ta</td><td align="left" valign="bottom">constant Wright-Fisher</td><td align="left" valign="bottom">Driver</td><td align="left" valign="bottom">0.233</td><td align="left" valign="bottom">1</td><td align="left" valign="bottom">0.0023 ± 1e-04</td><td align="left" valign="bottom">1 ± 0</td><td align="left" valign="bottom">57.8185 ± 0.6927</td></tr><tr><td align="left" valign="bottom">19Ta</td><td align="left" valign="bottom">constant Wright-Fisher</td><td align="left" valign="bottom">Hybrid</td><td align="left" valign="bottom">0.261</td><td align="left" valign="bottom">1</td><td align="left" valign="bottom">0.0023 ± 1e-04</td><td align="left" valign="bottom">1 ± 0</td><td align="left" valign="bottom">58.0478 ± 0.6705</td></tr><tr><td align="left" valign="bottom">19Ta</td><td align="left" valign="bottom">constant Wright-Fisher</td><td align="left" valign="bottom">Neutral</td><td align="left" valign="bottom">0.237</td><td align="left" valign="bottom">1</td><td align="left" valign="bottom">0.0025 ± 0.00012</td><td align="left" valign="bottom">1 ± 0</td><td align="left" valign="bottom">57.2261 ± 0.6669</td></tr><tr><td align="left" valign="bottom">19Tb</td><td align="left" valign="bottom">exponential pseudo-Moran</td><td align="left" valign="bottom">Abundance</td><td align="left" valign="bottom">0.275</td><td align="left" valign="bottom">1</td><td align="left" valign="bottom">9e-04 ± 5e-05</td><td align="left" valign="bottom">1 ± 0</td><td align="left" valign="bottom">58.1624 ± 0.6643</td></tr><tr><td align="left" valign="bottom">19Tb</td><td align="left" valign="bottom">exponential pseudo-Moran</td><td align="left" valign="bottom">Driver</td><td align="left" valign="bottom">0.239</td><td align="left" valign="bottom">1</td><td align="left" valign="bottom">9e-04 ± 5e-05</td><td align="left" valign="bottom">1 ± 0</td><td align="left" valign="bottom">58.4554 ± 0.6736</td></tr><tr><td align="left" valign="bottom">19Tb</td><td align="left" valign="bottom">exponential pseudo-Moran</td><td align="left" valign="bottom">Hybrid</td><td align="left" valign="bottom">0.258</td><td align="left" valign="bottom">1</td><td align="left" valign="bottom">9e-04 ± 5e-05</td><td align="left" valign="bottom">1 ± 0</td><td align="left" valign="bottom">58.586 ± 0.6668</td></tr><tr><td align="left" valign="bottom">19Tb</td><td align="left" valign="bottom">exponential pseudo-Moran</td><td align="left" valign="bottom">Neutral</td><td align="left" valign="bottom">0.228</td><td align="left" valign="bottom">1</td><td align="left" valign="bottom">9e-04 ± 6e-05</td><td align="left" valign="bottom">1 ± 0</td><td align="left" valign="bottom">58.5796 ± 0.6796</td></tr><tr><td align="left" valign="bottom">19Tb</td><td align="left" valign="bottom">constant Wright-Fisher</td><td align="left" valign="bottom">Abundance</td><td align="left" valign="bottom">0.259</td><td align="left" valign="bottom">1</td><td align="left" valign="bottom">9e-04 ± 6e-05</td><td align="left" valign="bottom">1 ± 0</td><td align="left" valign="bottom">58.7611 ± 0.6614</td></tr><tr><td align="left" valign="bottom">19Tb</td><td align="left" valign="bottom">constant Wright-Fisher</td><td align="left" valign="bottom">Driver</td><td align="left" valign="bottom">0.24</td><td align="left" valign="bottom">1</td><td align="left" valign="bottom">9e-04 ± 6e-05</td><td align="left" valign="bottom">1 ± 0</td><td align="left" valign="bottom">58.1178 ± 0.679</td></tr><tr><td align="left" valign="bottom">19Tb</td><td align="left" valign="bottom">constant Wright-Fisher</td><td align="left" valign="bottom">Hybrid</td><td align="left" valign="bottom">0.257</td><td align="left" valign="bottom">1</td><td align="left" valign="bottom">9e-04 ± 5e-05</td><td align="left" valign="bottom">1 ± 0</td><td align="left" valign="bottom">59.1592 ± 0.6715</td></tr><tr><td align="left" valign="bottom">19Tb</td><td align="left" valign="bottom">constant Wright-Fisher</td><td align="left" valign="bottom">Neutral</td><td align="left" valign="bottom">0.245</td><td align="left" valign="bottom">1</td><td align="left" valign="bottom">9e-04 ± 7e-05</td><td align="left" valign="bottom">1 ± 0</td><td align="left" valign="bottom">58.7516 ± 0.6787</td></tr><tr><td align="left" valign="bottom">24Ta</td><td align="left" valign="bottom">exponential pseudo-Moran</td><td align="left" valign="bottom">Abundance</td><td align="left" valign="bottom">0.275</td><td align="left" valign="bottom">1</td><td align="left" valign="bottom">9e-04 ± 5e-05</td><td align="left" valign="bottom">1 ± 0</td><td align="left" valign="bottom">58.1624 ± 0.6643</td></tr><tr><td align="left" valign="bottom">24Ta</td><td align="left" valign="bottom">exponential pseudo-Moran</td><td align="left" valign="bottom">Driver</td><td align="left" valign="bottom">0.239</td><td align="left" valign="bottom">1</td><td align="left" valign="bottom">9e-04 ± 5e-05</td><td align="left" valign="bottom">1 ± 0</td><td align="left" valign="bottom">58.4554 ± 0.6736</td></tr><tr><td align="left" valign="bottom">24Ta</td><td align="left" valign="bottom">exponential pseudo-Moran</td><td align="left" valign="bottom">Hybrid</td><td align="left" valign="bottom">0.258</td><td align="left" valign="bottom">1</td><td align="left" valign="bottom">9e-04 ± 5e-05</td><td align="left" valign="bottom">1 ± 0</td><td align="left" valign="bottom">58.586 ± 0.6668</td></tr><tr><td align="left" valign="bottom">24Ta</td><td align="left" valign="bottom">exponential pseudo-Moran</td><td align="left" valign="bottom">Neutral</td><td align="left" valign="bottom">0.228</td><td align="left" valign="bottom">1</td><td align="left" valign="bottom">9e-04 ± 6e-05</td><td align="left" valign="bottom">1 ± 0</td><td align="left" valign="bottom">58.6656 ± 0.6783</td></tr><tr><td align="left" valign="bottom">24Ta</td><td align="left" valign="bottom">constant Wright-Fisher</td><td align="left" valign="bottom">Abundance</td><td align="left" valign="bottom">0.259</td><td align="left" valign="bottom">1</td><td align="left" valign="bottom">9e-04 ± 6e-05</td><td align="left" valign="bottom">1 ± 0</td><td align="left" valign="bottom">58.7611 ± 0.6614</td></tr><tr><td align="left" valign="bottom">24Ta</td><td align="left" valign="bottom">constant Wright-Fisher</td><td align="left" valign="bottom">Driver</td><td align="left" valign="bottom">0.24</td><td align="left" valign="bottom">1</td><td align="left" valign="bottom">9e-04 ± 6e-05</td><td align="left" valign="bottom">1 ± 0</td><td align="left" valign="bottom">58.1783 ± 0.6771</td></tr><tr><td align="left" valign="bottom">24Ta</td><td align="left" valign="bottom">constant Wright-Fisher</td><td align="left" valign="bottom">Hybrid</td><td align="left" valign="bottom">0.257</td><td align="left" valign="bottom">1</td><td align="left" valign="bottom">9e-04 ± 5e-05</td><td align="left" valign="bottom">1 ± 0</td><td align="left" valign="bottom">59.1592 ± 0.6715</td></tr><tr><td align="left" valign="bottom">24Ta</td><td align="left" valign="bottom">constant Wright-Fisher</td><td align="left" valign="bottom">Neutral</td><td align="left" valign="bottom">0.245</td><td align="left" valign="bottom">1</td><td align="left" valign="bottom">9e-04 ± 7e-05</td><td align="left" valign="bottom">1 ± 0</td><td align="left" valign="bottom">58.7516 ± 0.6787</td></tr><tr><td align="left" valign="bottom">24Tb</td><td align="left" valign="bottom">exponential pseudo-Moran</td><td align="left" valign="bottom">Abundance</td><td align="left" valign="bottom">0.273</td><td align="left" valign="bottom">1</td><td align="left" valign="bottom">0.0023 ± 0.00011</td><td align="left" valign="bottom">1 ± 0</td><td align="left" valign="bottom">57.0446 ± 0.6526</td></tr><tr><td align="left" valign="bottom">24Tb</td><td align="left" valign="bottom">exponential pseudo-Moran</td><td align="left" valign="bottom">Driver</td><td align="left" valign="bottom">0.242</td><td align="left" valign="bottom">1</td><td align="left" valign="bottom">0.0025 ± 0.00012</td><td align="left" valign="bottom">1 ± 0</td><td align="left" valign="bottom">57.551 ± 0.6661</td></tr><tr><td align="left" valign="bottom">24Tb</td><td align="left" valign="bottom">exponential pseudo-Moran</td><td align="left" valign="bottom">Hybrid</td><td align="left" valign="bottom">0.264</td><td align="left" valign="bottom">1</td><td align="left" valign="bottom">0.0022 ± 9e-05</td><td align="left" valign="bottom">1 ± 0</td><td align="left" valign="bottom">57.9108 ± 0.6512</td></tr><tr><td align="left" valign="bottom">24Tb</td><td align="left" valign="bottom">exponential pseudo-Moran</td><td align="left" valign="bottom">Neutral</td><td align="left" valign="bottom">0.222</td><td align="left" valign="bottom">1</td><td align="left" valign="bottom">0.0026 ± 0.00013</td><td align="left" valign="bottom">1 ± 0</td><td align="left" valign="bottom">57.7516 ± 0.6758</td></tr><tr><td align="left" valign="bottom">24Tb</td><td align="left" valign="bottom">constant Wright-Fisher</td><td align="left" valign="bottom">Abundance</td><td align="left" valign="bottom">0.267</td><td align="left" valign="bottom">1</td><td align="left" valign="bottom">0.0024 ± 0.00013</td><td align="left" valign="bottom">1 ± 0</td><td align="left" valign="bottom">58.379 ± 0.6601</td></tr><tr><td align="left" valign="bottom">24Tb</td><td align="left" valign="bottom">constant Wright-Fisher</td><td align="left" valign="bottom">Driver</td><td align="left" valign="bottom">0.237</td><td align="left" valign="bottom">1</td><td align="left" valign="bottom">0.0024 ± 1e-04</td><td align="left" valign="bottom">1 ± 0</td><td align="left" valign="bottom">57.7357 ± 0.6922</td></tr><tr><td align="left" valign="bottom">24Tb</td><td align="left" valign="bottom">constant Wright-Fisher</td><td align="left" valign="bottom">Hybrid</td><td align="left" valign="bottom">0.257</td><td align="left" valign="bottom">1</td><td align="left" valign="bottom">0.0023 ± 1e-04</td><td align="left" valign="bottom">1 ± 0</td><td align="left" valign="bottom">57.9045 ± 0.6718</td></tr><tr><td align="left" valign="bottom">24Tb</td><td align="left" valign="bottom">constant Wright-Fisher</td><td align="left" valign="bottom">Neutral</td><td align="left" valign="bottom">0.239</td><td align="left" valign="bottom">1</td><td align="left" valign="bottom">0.0025 ± 0.00012</td><td align="left" valign="bottom">1 ± 0</td><td align="left" valign="bottom">57.2643 ± 0.6726</td></tr><tr><td align="left" valign="bottom">26N</td><td align="left" valign="bottom">exponential pseudo-Moran</td><td align="left" valign="bottom">Abundance</td><td align="left" valign="bottom">0.274</td><td align="left" valign="bottom">1</td><td align="left" valign="bottom">9e-04 ± 5e-05</td><td align="left" valign="bottom">1 ± 0</td><td align="left" valign="bottom">58.2452 ± 0.6646</td></tr><tr><td align="left" valign="bottom">26N</td><td align="left" valign="bottom">exponential pseudo-Moran</td><td align="left" valign="bottom">Driver</td><td align="left" valign="bottom">0.239</td><td align="left" valign="bottom">1</td><td align="left" valign="bottom">9e-04 ± 5e-05</td><td align="left" valign="bottom">1 ± 0</td><td align="left" valign="bottom">58.4045 ± 0.6706</td></tr><tr><td align="left" valign="bottom">26N</td><td align="left" valign="bottom">exponential pseudo-Moran</td><td align="left" valign="bottom">Hybrid</td><td align="left" valign="bottom">0.26</td><td align="left" valign="bottom">1</td><td align="left" valign="bottom">9e-04 ± 5e-05</td><td align="left" valign="bottom">1 ± 0</td><td align="left" valign="bottom">58.586 ± 0.6668</td></tr><tr><td align="left" valign="bottom">26N</td><td align="left" valign="bottom">exponential pseudo-Moran</td><td align="left" valign="bottom">Neutral</td><td align="left" valign="bottom">0.227</td><td align="left" valign="bottom">1</td><td align="left" valign="bottom">0.001 ± 7e-05</td><td align="left" valign="bottom">1 ± 0</td><td align="left" valign="bottom">58.6815 ± 0.6776</td></tr><tr><td align="left" valign="bottom">26N</td><td align="left" valign="bottom">constant Wright-Fisher</td><td align="left" valign="bottom">Abundance</td><td align="left" valign="bottom">0.259</td><td align="left" valign="bottom">1</td><td align="left" valign="bottom">9e-04 ± 6e-05</td><td align="left" valign="bottom">1 ± 0</td><td align="left" valign="bottom">58.8089 ± 0.6627</td></tr><tr><td align="left" valign="bottom">26N</td><td align="left" valign="bottom">constant Wright-Fisher</td><td align="left" valign="bottom">Driver</td><td align="left" valign="bottom">0.239</td><td align="left" valign="bottom">1</td><td align="left" valign="bottom">9e-04 ± 6e-05</td><td align="left" valign="bottom">1 ± 0</td><td align="left" valign="bottom">58.1783 ± 0.6771</td></tr><tr><td align="left" valign="bottom">26N</td><td align="left" valign="bottom">constant Wright-Fisher</td><td align="left" valign="bottom">Hybrid</td><td align="left" valign="bottom">0.257</td><td align="left" valign="bottom">1</td><td align="left" valign="bottom">9e-04 ± 5e-05</td><td align="left" valign="bottom">1 ± 0</td><td align="left" valign="bottom">59.1178 ± 0.6745</td></tr><tr><td align="left" valign="bottom">26N</td><td align="left" valign="bottom">constant Wright-Fisher</td><td align="left" valign="bottom">Neutral</td><td align="left" valign="bottom">0.245</td><td align="left" valign="bottom">1</td><td align="left" valign="bottom">0.001 ± 7e-05</td><td align="left" valign="bottom">1 ± 0</td><td align="left" valign="bottom">58.6879 ± 0.6762</td></tr><tr><td align="left" valign="bottom">9T</td><td align="left" valign="bottom">exponential pseudo-Moran</td><td align="left" valign="bottom">Abundance</td><td align="left" valign="bottom">0.274</td><td align="left" valign="bottom">1</td><td align="left" valign="bottom">0.0021 ± 8e-05</td><td align="left" valign="bottom">1 ± 0</td><td align="left" valign="bottom">57.3854 ± 0.6574</td></tr><tr><td align="left" valign="bottom">9T</td><td align="left" valign="bottom">exponential pseudo-Moran</td><td align="left" valign="bottom">Driver</td><td align="left" valign="bottom">0.242</td><td align="left" valign="bottom">1</td><td align="left" valign="bottom">0.0024 ± 0.00011</td><td align="left" valign="bottom">1 ± 0</td><td align="left" valign="bottom">57.8025 ± 0.663</td></tr><tr><td align="left" valign="bottom">9T</td><td align="left" valign="bottom">exponential pseudo-Moran</td><td align="left" valign="bottom">Hybrid</td><td align="left" valign="bottom">0.261</td><td align="left" valign="bottom">1</td><td align="left" valign="bottom">0.0022 ± 8e-05</td><td align="left" valign="bottom">1 ± 0</td><td align="left" valign="bottom">57.9108 ± 0.65</td></tr><tr><td align="left" valign="bottom">9T</td><td align="left" valign="bottom">exponential pseudo-Moran</td><td align="left" valign="bottom">Neutral</td><td align="left" valign="bottom">0.222</td><td align="left" valign="bottom">1</td><td align="left" valign="bottom">0.0025 ± 0.00012</td><td align="left" valign="bottom">1 ± 0</td><td align="left" valign="bottom">57.9522 ± 0.6787</td></tr><tr><td align="left" valign="bottom">9T</td><td align="left" valign="bottom">constant Wright-Fisher</td><td align="left" valign="bottom">Abundance</td><td align="left" valign="bottom">0.269</td><td align="left" valign="bottom">1</td><td align="left" valign="bottom">0.0024 ± 0.00011</td><td align="left" valign="bottom">1 ± 0</td><td align="left" valign="bottom">58.2866 ± 0.6566</td></tr><tr><td align="left" valign="bottom">9T</td><td align="left" valign="bottom">constant Wright-Fisher</td><td align="left" valign="bottom">Driver</td><td align="left" valign="bottom">0.233</td><td align="left" valign="bottom">1</td><td align="left" valign="bottom">0.0023 ± 1e-04</td><td align="left" valign="bottom">1 ± 0</td><td align="left" valign="bottom">57.9076 ± 0.6927</td></tr><tr><td align="left" valign="bottom">9T</td><td align="left" valign="bottom">constant Wright-Fisher</td><td align="left" valign="bottom">Hybrid</td><td align="left" valign="bottom">0.261</td><td align="left" valign="bottom">1</td><td align="left" valign="bottom">0.0023 ± 1e-04</td><td align="left" valign="bottom">1 ± 0</td><td align="left" valign="bottom">58.1115 ± 0.6708</td></tr><tr><td align="left" valign="bottom">9T</td><td align="left" valign="bottom">constant Wright-Fisher</td><td align="left" valign="bottom">Neutral</td><td align="left" valign="bottom">0.236</td><td align="left" valign="bottom">1</td><td align="left" valign="bottom">0.0025 ± 0.00012</td><td align="left" valign="bottom">1 ± 0</td><td align="left" valign="bottom">57.2261 ± 0.6669</td></tr><tr><td align="left" valign="bottom">PolyB1</td><td align="left" valign="bottom">exponential pseudo-Moran</td><td align="left" valign="bottom">Abundance</td><td align="left" valign="bottom">0.274</td><td align="left" valign="bottom">1</td><td align="left" valign="bottom">0.0021 ± 8e-05</td><td align="left" valign="bottom">1 ± 0</td><td align="left" valign="bottom">57.4045 ± 0.6565</td></tr><tr><td align="left" valign="bottom">PolyB1</td><td align="left" valign="bottom">exponential pseudo-Moran</td><td align="left" valign="bottom">Driver</td><td align="left" valign="bottom">0.243</td><td align="left" valign="bottom">1</td><td align="left" valign="bottom">0.0024 ± 0.00011</td><td align="left" valign="bottom">1 ± 0</td><td align="left" valign="bottom">57.7102 ± 0.6622</td></tr><tr><td align="left" valign="bottom">PolyB1</td><td align="left" valign="bottom">exponential pseudo-Moran</td><td align="left" valign="bottom">Hybrid</td><td align="left" valign="bottom">0.261</td><td align="left" valign="bottom">1</td><td align="left" valign="bottom">0.0022 ± 8e-05</td><td align="left" valign="bottom">1 ± 0</td><td align="left" valign="bottom">57.9459 ± 0.6512</td></tr><tr><td align="left" valign="bottom">PolyB1</td><td align="left" valign="bottom">exponential pseudo-Moran</td><td align="left" valign="bottom">Neutral</td><td align="left" valign="bottom">0.222</td><td align="left" valign="bottom">1</td><td align="left" valign="bottom">0.0025 ± 0.00011</td><td align="left" valign="bottom">1 ± 0</td><td align="left" valign="bottom">57.9522 ± 0.6776</td></tr><tr><td align="left" valign="bottom">PolyB1</td><td align="left" valign="bottom">constant Wright-Fisher</td><td align="left" valign="bottom">Abundance</td><td align="left" valign="bottom">0.271</td><td align="left" valign="bottom">1</td><td align="left" valign="bottom">0.0023 ± 0.00011</td><td align="left" valign="bottom">1 ± 0</td><td align="left" valign="bottom">58.2834 ± 0.6575</td></tr><tr><td align="left" valign="bottom">PolyB1</td><td align="left" valign="bottom">constant Wright-Fisher</td><td align="left" valign="bottom">Driver</td><td align="left" valign="bottom">0.231</td><td align="left" valign="bottom">1</td><td align="left" valign="bottom">0.0023 ± 9e-05</td><td align="left" valign="bottom">1 ± 0</td><td align="left" valign="bottom">57.6656 ± 0.6949</td></tr><tr><td align="left" valign="bottom">PolyB1</td><td align="left" valign="bottom">constant Wright-Fisher</td><td align="left" valign="bottom">Hybrid</td><td align="left" valign="bottom">0.261</td><td align="left" valign="bottom">1</td><td align="left" valign="bottom">0.0023 ± 1e-04</td><td align="left" valign="bottom">1 ± 0</td><td align="left" valign="bottom">57.9713 ± 0.6668</td></tr><tr><td align="left" valign="bottom">PolyB1</td><td align="left" valign="bottom">constant Wright-Fisher</td><td align="left" valign="bottom">Neutral</td><td align="left" valign="bottom">0.237</td><td align="left" valign="bottom">1</td><td align="left" valign="bottom">0.0025 ± 0.00012</td><td align="left" valign="bottom">1 ± 0</td><td align="left" valign="bottom">57.207 ± 0.6674</td></tr><tr><td align="left" valign="bottom">PolyB2</td><td align="left" valign="bottom">exponential pseudo-Moran</td><td align="left" valign="bottom">Abundance</td><td align="left" valign="bottom">0.272</td><td align="left" valign="bottom">1</td><td align="left" valign="bottom">0.0027 ± 2e-04</td><td align="left" valign="bottom">1 ± 0</td><td align="left" valign="bottom">56.8471 ± 0.6544</td></tr><tr><td align="left" valign="bottom">PolyB2</td><td align="left" valign="bottom">exponential pseudo-Moran</td><td align="left" valign="bottom">Driver</td><td align="left" valign="bottom">0.245</td><td align="left" valign="bottom">1</td><td align="left" valign="bottom">0.0029 ± 0.00021</td><td align="left" valign="bottom">1 ± 0</td><td align="left" valign="bottom">57.3312 ± 0.6609</td></tr><tr><td align="left" valign="bottom">PolyB2</td><td align="left" valign="bottom">exponential pseudo-Moran</td><td align="left" valign="bottom">Hybrid</td><td align="left" valign="bottom">0.263</td><td align="left" valign="bottom">1</td><td align="left" valign="bottom">0.0024 ± 0.00011</td><td align="left" valign="bottom">1 ± 0</td><td align="left" valign="bottom">57.9204 ± 0.6466</td></tr><tr><td align="left" valign="bottom">PolyB2</td><td align="left" valign="bottom">exponential pseudo-Moran</td><td align="left" valign="bottom">Neutral</td><td align="left" valign="bottom">0.221</td><td align="left" valign="bottom">1</td><td align="left" valign="bottom">0.0029 ± 0.00017</td><td align="left" valign="bottom">1 ± 0</td><td align="left" valign="bottom">57.4236 ± 0.6784</td></tr><tr><td align="left" valign="bottom">PolyB2</td><td align="left" valign="bottom">constant Wright-Fisher</td><td align="left" valign="bottom">Abundance</td><td align="left" valign="bottom">0.268</td><td align="left" valign="bottom">1</td><td align="left" valign="bottom">0.0025 ± 0.00013</td><td align="left" valign="bottom">1 ± 0</td><td align="left" valign="bottom">58.2484 ± 0.6616</td></tr><tr><td align="left" valign="bottom">PolyB2</td><td align="left" valign="bottom">constant Wright-Fisher</td><td align="left" valign="bottom">Driver</td><td align="left" valign="bottom">0.235</td><td align="left" valign="bottom">1</td><td align="left" valign="bottom">0.0026 ± 0.00014</td><td align="left" valign="bottom">1 ± 0</td><td align="left" valign="bottom">57.5796 ± 0.6897</td></tr><tr><td align="left" valign="bottom">PolyB2</td><td align="left" valign="bottom">constant Wright-Fisher</td><td align="left" valign="bottom">Hybrid</td><td align="left" valign="bottom">0.257</td><td align="left" valign="bottom">1</td><td align="left" valign="bottom">0.0026 ± 0.00015</td><td align="left" valign="bottom">1 ± 0</td><td align="left" valign="bottom">58.1115 ± 0.6741</td></tr><tr><td align="left" valign="bottom">PolyB2</td><td align="left" valign="bottom">constant Wright-Fisher</td><td align="left" valign="bottom">Neutral</td><td align="left" valign="bottom">0.24</td><td align="left" valign="bottom">1</td><td align="left" valign="bottom">0.0027 ± 0.00014</td><td align="left" valign="bottom">1 ± 0</td><td align="left" valign="bottom">57.379 ± 0.6701</td></tr></tbody></table></table-wrap><table-wrap id="table5" position="float"><label>Table 5.</label><caption><title>Approximate reported per chromosome mis-segregation rates.</title></caption><table frame="hsides" rules="groups"><thead><tr><th align="char" char="." valign="bottom">1st Author</th><th align="left" valign="bottom">DOI</th><th align="left" valign="bottom">Model</th><th align="left" valign="bottom">Tumor?</th><th align="left" valign="bottom">Statistic</th><th align="left" valign="bottom">Assessment</th><th align="left" valign="bottom">Approximate observed frequency %</th><th align="left" valign="bottom">Aprrox modal chromosome # (ATCC)</th><th align="left" valign="bottom">Approximate mis-segregation rate (per chromosome)</th></tr></thead><tbody><tr><td align="left" valign="bottom">Bakhoum</td><td align="left" valign="bottom"><ext-link ext-link-type="uri" xlink:href="https://doi.org/10.1158/1078-0432.CCR-11-2049">https://doi.org/10.1158/1078-0432.CCR-11-2049</ext-link></td><td align="left" valign="bottom">Tumor-TMA</td><td align="left" valign="bottom">Tumor</td><td align="left" valign="bottom">Reported</td><td align="left" valign="bottom">Lagging/Bridging</td><td align="char" char="." valign="bottom">31.3</td><td align="char" char="." valign="bottom">46</td><td align="char" char="." valign="bottom">0.00680</td></tr><tr><td align="left" valign="bottom">Orr</td><td align="left" valign="bottom"><ext-link ext-link-type="uri" xlink:href="https://doi.org/10.1016/j.celrep.2016.10.030">https://doi.org/10.1016/j.celrep.2016.10.030</ext-link></td><td align="left" valign="bottom">U2OS</td><td align="left" valign="bottom">Tumor</td><td align="left" valign="bottom">Approx. Mean</td><td align="left" valign="bottom">Lagging</td><td align="char" char="." valign="bottom">32.5</td><td align="char" char="." valign="bottom">46</td><td align="char" char="." valign="bottom">0.00707</td></tr><tr><td align="left" valign="bottom">Orr</td><td align="left" valign="bottom"><ext-link ext-link-type="uri" xlink:href="https://doi.org/10.1016/j.celrep.2016.10.030">https://doi.org/10.1016/j.celrep.2016.10.030</ext-link></td><td align="left" valign="bottom">HeLa</td><td align="left" valign="bottom">Tumor</td><td align="left" valign="bottom">Approx. Mean</td><td align="left" valign="bottom">Lagging</td><td align="char" char="." valign="bottom">22</td><td align="char" char="." valign="bottom">82</td><td align="char" char="." valign="bottom">0.00268</td></tr><tr><td align="left" valign="bottom">Orr</td><td align="left" valign="bottom"><ext-link ext-link-type="uri" xlink:href="https://doi.org/10.1016/j.celrep.2016.10.030">https://doi.org/10.1016/j.celrep.2016.10.030</ext-link></td><td align="left" valign="bottom">SW-620</td><td align="left" valign="bottom">Tumor</td><td align="left" valign="bottom">Approx. Mean</td><td align="left" valign="bottom">Lagging</td><td align="char" char="." valign="bottom">22.5</td><td align="char" char="." valign="bottom">50</td><td align="char" char="." valign="bottom">0.00450</td></tr><tr><td align="left" valign="bottom">Orr</td><td align="left" valign="bottom"><ext-link ext-link-type="uri" xlink:href="https://doi.org/10.1016/j.celrep.2016.10.030">https://doi.org/10.1016/j.celrep.2016.10.030</ext-link></td><td align="left" valign="bottom">RPE1</td><td align="left" valign="bottom">Non-tumor</td><td align="left" valign="bottom">Approx. Mean</td><td align="left" valign="bottom">Lagging</td><td align="char" char="." valign="bottom">2.5</td><td align="char" char="." valign="bottom">46</td><td align="char" char="." valign="bottom">0.00054</td></tr><tr><td align="left" valign="bottom">Orr</td><td align="left" valign="bottom"><ext-link ext-link-type="uri" xlink:href="https://doi.org/10.1016/j.celrep.2016.10.030">https://doi.org/10.1016/j.celrep.2016.10.030</ext-link></td><td align="left" valign="bottom">BJ</td><td align="left" valign="bottom">Non-tumor</td><td align="left" valign="bottom">Approx. Mean</td><td align="left" valign="bottom">Lagging</td><td align="char" char="." valign="bottom">8</td><td align="char" char="." valign="bottom">46</td><td align="char" char="." valign="bottom">0.00174</td></tr><tr><td align="left" valign="bottom">Nicholson</td><td align="left" valign="bottom"><ext-link ext-link-type="uri" xlink:href="https://doi.org/10.7554/eLife.05068">https://doi.org/10.7554/eLife.05068</ext-link></td><td align="left" valign="bottom">Amniocyte</td><td align="left" valign="bottom">Non-tumor</td><td align="left" valign="bottom">Approx. Mean</td><td align="left" valign="bottom">Lagging</td><td align="char" char="." valign="bottom">0</td><td align="char" char="." valign="bottom">46</td><td align="char" char="." valign="bottom">0.00000</td></tr><tr><td align="left" valign="bottom">Nicholson</td><td align="left" valign="bottom"><ext-link ext-link-type="uri" xlink:href="https://doi.org/10.7554/eLife.05068">https://doi.org/10.7554/eLife.05068</ext-link></td><td align="left" valign="bottom">DLD1</td><td align="left" valign="bottom">Tumor</td><td align="left" valign="bottom">Approx. Mean</td><td align="left" valign="bottom">Lagging</td><td align="char" char="." valign="bottom">1</td><td align="char" char="." valign="bottom">46</td><td align="char" char="." valign="bottom">0.00022</td></tr><tr><td align="left" valign="bottom">Dewhurst</td><td align="left" valign="bottom"><ext-link ext-link-type="uri" xlink:href="https://doi.org/10.1158/2159-8290.CD-13-0285">https://doi.org/10.1158/2159-8290.CD-13-0285</ext-link></td><td align="left" valign="bottom">HCT116-Diploid</td><td align="left" valign="bottom">Tumor</td><td align="left" valign="bottom">Approx. Mean</td><td align="left" valign="bottom">Lagging/Bridging</td><td align="char" char="." valign="bottom">23</td><td align="char" char="." valign="bottom">45</td><td align="char" char="." valign="bottom">0.00511</td></tr><tr><td align="left" valign="bottom">Dewhurst</td><td align="left" valign="bottom"><ext-link ext-link-type="uri" xlink:href="https://doi.org/10.1158/2159-8290.CD-13-0285">https://doi.org/10.1158/2159-8290.CD-13-0285</ext-link></td><td align="left" valign="bottom">HCT116-Tetraploid</td><td align="left" valign="bottom">Tumor</td><td align="left" valign="bottom">Approx. Mean</td><td align="left" valign="bottom">Lagging/Bridging</td><td align="char" char="." valign="bottom">50</td><td align="char" char="." valign="bottom">90</td><td align="char" char="." valign="bottom">0.00556</td></tr><tr><td align="left" valign="bottom">Bakhoum</td><td align="left" valign="bottom"><ext-link ext-link-type="uri" xlink:href="https://doi.org/10.1038/ncb1809">https://doi.org/10.1038/ncb1809</ext-link></td><td align="left" valign="bottom">U2OS</td><td align="left" valign="bottom">Tumor</td><td align="left" valign="bottom">Reported</td><td align="left" valign="bottom">Lagging</td><td align="left" valign="bottom"/><td align="char" char="." valign="bottom">46</td><td align="char" char="." valign="bottom">0.01000</td></tr><tr><td align="left" valign="bottom">Zasadil</td><td align="left" valign="bottom"><ext-link ext-link-type="uri" xlink:href="https://doi.org/10.1126/scitranslmed.3007965">https://doi.org/10.1126/scitranslmed.3007965</ext-link></td><td align="left" valign="bottom">CAL51</td><td align="left" valign="bottom">Tumor</td><td align="left" valign="bottom">Approx. Mean</td><td align="left" valign="bottom">Lagging</td><td align="char" char="." valign="bottom">0.5</td><td align="char" char="." valign="bottom">44</td><td align="char" char="." valign="bottom">0.00011</td></tr><tr><td align="left" valign="bottom">Thompson</td><td align="left" valign="bottom"><ext-link ext-link-type="uri" xlink:href="https://doi.org/10.1083/jcb.200712029">https://doi.org/10.1083/jcb.200712029</ext-link></td><td align="left" valign="bottom">RPE1</td><td align="left" valign="bottom">Non-tumor</td><td align="left" valign="bottom">Approx. Mean</td><td align="left" colspan="2" valign="bottom">Acute aneuploidy via FISH</td><td align="char" char="." valign="bottom">46</td><td align="char" char="." valign="bottom">0.00025</td></tr><tr><td align="left" valign="bottom">Thompson</td><td align="left" valign="bottom"><ext-link ext-link-type="uri" xlink:href="https://doi.org/10.1083/jcb.200712029">https://doi.org/10.1083/jcb.200712029</ext-link></td><td align="left" valign="bottom">HCT116-Diploid</td><td align="left" valign="bottom">Tumor</td><td align="left" valign="bottom">Approx. Mean</td><td align="left" colspan="2" valign="bottom">Acute aneuploidy via FISH</td><td align="char" char="." valign="bottom">45</td><td align="char" char="." valign="bottom">0.00025</td></tr><tr><td align="left" valign="bottom">Thompson</td><td align="left" valign="bottom"><ext-link ext-link-type="uri" xlink:href="https://doi.org/10.1083/jcb.200712029">https://doi.org/10.1083/jcb.200712029</ext-link></td><td align="left" valign="bottom">HT29</td><td align="left" valign="bottom">Tumor</td><td align="left" valign="bottom">Approx. Mean</td><td align="left" colspan="2" valign="bottom">Acute aneuploidy via FISH</td><td align="char" char="." valign="bottom">71</td><td align="char" char="." valign="bottom">0.00250</td></tr><tr><td align="left" valign="bottom">Thompson</td><td align="left" valign="bottom"><ext-link ext-link-type="uri" xlink:href="https://doi.org/10.1083/jcb.200712029">https://doi.org/10.1083/jcb.200712029</ext-link></td><td align="left" valign="bottom">Caco2</td><td align="left" valign="bottom">Tumor</td><td align="left" valign="bottom">Approx. Mean</td><td align="left" colspan="2" valign="bottom">Acute aneuploidy via FISH</td><td align="char" char="." valign="bottom">96</td><td align="char" char="." valign="bottom">0.00900</td></tr><tr><td align="left" valign="bottom">Thompson</td><td align="left" valign="bottom"><ext-link ext-link-type="uri" xlink:href="https://doi.org/10.1083/jcb.200712029">https://doi.org/10.1083/jcb.200712029</ext-link></td><td align="left" valign="bottom">MCF-7</td><td align="left" valign="bottom">Tumor</td><td align="left" valign="bottom">Approx. Mean</td><td align="left" colspan="2" valign="bottom">Acute aneuploidy via FISH</td><td align="char" char="." valign="bottom">82</td><td align="char" char="." valign="bottom">0.00700</td></tr><tr><td align="left" valign="bottom">Bakhoum</td><td align="left" valign="bottom"><ext-link ext-link-type="uri" xlink:href="https://doi.org/10.1016/j.cub.2014.01.019">https://doi.org/10.1016/j.cub.2014.01.019</ext-link></td><td align="left" valign="bottom">HCT116-Diploid</td><td align="left" valign="bottom">Tumor</td><td align="left" valign="bottom">Approx. Mean</td><td align="left" valign="bottom">Lagging</td><td align="char" char="." valign="bottom">6</td><td align="char" char="." valign="bottom">45</td><td align="char" char="." valign="bottom">0.00133</td></tr><tr><td align="left" valign="bottom">Bakhoum</td><td align="left" valign="bottom"><ext-link ext-link-type="uri" xlink:href="https://doi.org/10.1016/j.cub.2014.01.019">https://doi.org/10.1016/j.cub.2014.01.019</ext-link></td><td align="left" valign="bottom">DLD1</td><td align="left" valign="bottom">Tumor</td><td align="left" valign="bottom">Approx. Mean</td><td align="left" valign="bottom">Lagging</td><td align="char" char="." valign="bottom">2</td><td align="char" char="." valign="bottom">46</td><td align="char" char="." valign="bottom">0.00043</td></tr><tr><td align="left" valign="bottom">Bakhoum</td><td align="left" valign="bottom"><ext-link ext-link-type="uri" xlink:href="https://doi.org/10.1016/j.cub.2014.01.019">https://doi.org/10.1016/j.cub.2014.01.019</ext-link></td><td align="left" valign="bottom">HT29</td><td align="left" valign="bottom">Tumor</td><td align="left" valign="bottom">Approx. Mean</td><td align="left" valign="bottom">Lagging</td><td align="char" char="." valign="bottom">14</td><td align="char" char="." valign="bottom">71</td><td align="char" char="." valign="bottom">0.00197</td></tr><tr><td align="left" valign="bottom">Bakhoum</td><td align="left" valign="bottom"><ext-link ext-link-type="uri" xlink:href="https://doi.org/10.1016/j.cub.2014.01.019">https://doi.org/10.1016/j.cub.2014.01.019</ext-link></td><td align="left" valign="bottom">SW-620</td><td align="left" valign="bottom">Tumor</td><td align="left" valign="bottom">Approx. Mean</td><td align="left" valign="bottom">Lagging</td><td align="char" char="." valign="bottom">12</td><td align="char" char="." valign="bottom">50</td><td align="char" char="." valign="bottom">0.00240</td></tr><tr><td align="left" valign="bottom">Bakhoum</td><td align="left" valign="bottom"><ext-link ext-link-type="uri" xlink:href="https://doi.org/10.1016/j.cub.2014.01.019">https://doi.org/10.1016/j.cub.2014.01.019</ext-link></td><td align="left" valign="bottom">MCF-7</td><td align="left" valign="bottom">Tumor</td><td align="left" valign="bottom">Approx. Mean</td><td align="left" valign="bottom">Lagging</td><td align="char" char="." valign="bottom">17</td><td align="char" char="." valign="bottom">82</td><td align="char" char="." valign="bottom">0.00207</td></tr><tr><td align="left" valign="bottom">Bakhoum</td><td align="left" valign="bottom"><ext-link ext-link-type="uri" xlink:href="https://doi.org/10.1016/j.cub.2014.01.019">https://doi.org/10.1016/j.cub.2014.01.019</ext-link></td><td align="left" valign="bottom">HeLa</td><td align="left" valign="bottom">Tumor</td><td align="left" valign="bottom">Approx. Mean</td><td align="left" valign="bottom">Lagging</td><td align="char" char="." valign="bottom">13</td><td align="char" char="." valign="bottom">82</td><td align="char" char="." valign="bottom">0.00159</td></tr><tr><td align="left" valign="bottom">Worrall</td><td align="left" valign="bottom"><ext-link ext-link-type="uri" xlink:href="https://doi.org/10.1016/j.celrep.2018.05.047">https://doi.org/10.1016/j.celrep.2018.05.047</ext-link></td><td align="left" valign="bottom">BJ</td><td align="left" valign="bottom">Non-tumor</td><td align="left" valign="bottom">Approx. Mean</td><td align="left" valign="bottom">Unspecified Error</td><td align="char" char="." valign="bottom">5</td><td align="char" char="." valign="bottom">46</td><td align="char" char="." valign="bottom">0.00109</td></tr><tr><td align="left" valign="bottom">Worrall</td><td align="left" valign="bottom"><ext-link ext-link-type="uri" xlink:href="https://doi.org/10.1016/j.celrep.2018.05.047">https://doi.org/10.1016/j.celrep.2018.05.047</ext-link></td><td align="left" valign="bottom">RPE1</td><td align="left" valign="bottom">Non-tumor</td><td align="left" valign="bottom">Approx. Mean</td><td align="left" valign="bottom">Unspecified Error</td><td align="char" char="." valign="bottom">5</td><td align="char" char="." valign="bottom">46</td><td align="char" char="." valign="bottom">0.00109</td></tr></tbody></table></table-wrap><p>To confirm the relevance of the inferred scalar exponent we performed our model selection scheme using only the simulation data with unmodified fitness values (S = 1; <xref ref-type="table" rid="table4">Table 4</xref>). In this case, we found that the inferred mis-segregation rates for most samples fell well below the expected range found in cancer cell lines (<xref ref-type="fig" rid="fig6">Figure 6E</xref>). Additionally, when we inferred mis-segregation rates and selection in the early timepoint of longitudinally sequenced organoid clones from <xref ref-type="bibr" rid="bib8">Bolhaqueiro et al., 2019</xref>, the composition of the resultant populations simulated using these inferred characteristics better resembled the late-timepoint organoid data than those with unmodified selection values (S = 1; <xref ref-type="fig" rid="fig6s4">Figure 6—figure supplements 4</xref> and <xref ref-type="fig" rid="fig6s5">5</xref>).</p><p>As further validation for mis-segregation rates, we compared these inferred rates from CRC PDOs with those directly measured in live imaging from <xref ref-type="bibr" rid="bib8">Bolhaqueiro et al., 2019</xref>. Although mis-segregation cannot be directly inferred from microscopy, diversity should correlate with the observed rate of mitotic errors. There was a strong correlation but for two outliers—14T and U1T (<xref ref-type="fig" rid="fig6">Figure 6G</xref>). In fact, when adjusting to the same scale and correcting for cell ploidy, these data follow a strong positive linear trend with a slightly lower slope than a 1:1 correlation, which could reflect an overestimation of mis-segregation rates in the microscopy data (<xref ref-type="fig" rid="fig6">Figure 6H</xref>). Particularly with lagging chromosomes, despite a chromosome’s involvement in an observed segregation defect, it may end up in the correct daughter cell. Overall, these results indicate that the inferred measures using approximate Bayesian computation and scDNAseq account for selection and provide a quantitative measure of CIN.</p></sec></sec><sec id="s3" sec-type="discussion"><title>Discussion</title><p>The clinical assessment of mutations, short indels, and microsatellite instability in human cancer determined by short-read sequencing currently guide clinical care. By contrast, CIN is highly prevalent, yet has remained largely intractable to clinical measures. Single-cell DNA sequencing now promises detailed karyotypic analysis across hundreds of cells, yet selective pressure suppresses the observed karyotype heterogeneity within a tumor. Optimal clinical measurement of CIN may be achieved with scDNAseq, but must additionally account for selective pressure, which reduces karyotype heterogeneity.</p><p>Despite the major limitations with current measures of CIN, emerging evidence hints at its utility as a biomarker to predict benefit to cancer therapy. For example, CIN measures appear to predict therapeutic response to paclitaxel (<xref ref-type="bibr" rid="bib38">Janssen et al., 2009</xref>; <xref ref-type="bibr" rid="bib64">Scribano et al., 2021</xref>; <xref ref-type="bibr" rid="bib73">Swanton et al., 2009</xref>). Nevertheless, existing measures of CIN have had significant limitations. FISH and histological analysis of mitotic abnormalities are limited in quantifying specific chromosomes or requiring highly proliferative tumor types, such as lymphomas and leukemia. Gene expression profiles are proposed to correlate with CIN among populations of tumor samples (<xref ref-type="bibr" rid="bib10">Carter et al., 2006</xref>), although they happen to correlate better with tumor proliferation (<xref ref-type="bibr" rid="bib66">Sheltzer, 2013</xref>); in any case, they are correlations across populations of tumors, not suitable as an individualized diagnostic. We conclude that scDNAseq is the most complete and tractable measure of cellular karyotypes, and sampling at least 200 cells, coupled with computational models and ABC, promises to offer the best measure of tumor CIN.</p><p>Computational modeling of aneuploidy and CIN has been used to explore evolution in the context of numerical CIN and karyotype selection (<xref ref-type="bibr" rid="bib22">Elizalde et al., 2018</xref>; <xref ref-type="bibr" rid="bib27">Gao et al., 2016</xref>; <xref ref-type="bibr" rid="bib32">Gusev et al., 2001</xref>; <xref ref-type="bibr" rid="bib31">Gusev et al., 2000</xref>; <xref ref-type="bibr" rid="bib43">Laughney et al., 2015</xref>; <xref ref-type="bibr" rid="bib56">Nowak et al., 2002</xref>). Gusev and Nowak lay the foundation for mathematical modeling of CIN. While Gusev focused on the karyotypic outcomes of CIN, Nowak considered the effects of CIN-inducing mutations and the subsequent rate of LOH. Neither considered the individual fitness differences between specific karyotypes (<xref ref-type="bibr" rid="bib32">Gusev et al., 2001</xref>; <xref ref-type="bibr" rid="bib31">Gusev et al., 2000</xref>; <xref ref-type="bibr" rid="bib56">Nowak et al., 2002</xref>). This was improved in <xref ref-type="bibr" rid="bib43">Laughney et al., 2015</xref> and <xref ref-type="bibr" rid="bib22">Elizalde et al., 2018</xref> where the authors leveraged the chromosome scores derived in <xref ref-type="bibr" rid="bib15">Davoli et al., 2013</xref>, which enable the inclusion of oncogenes and tumor suppressors in models of CIN as we have done. These studies have provided important insights such as the role of whole-genome doubling as an evolutionary bridge to optimized chromosome stoichiometry. Yet the populations derived in these studies tend to vary to a greater degree than observed with scDNAseq, as they do not model strong selection against aneuploidy. Further, they do not attempt to use their models to measure CIN in biological samples. Here, we build on these models by considering, in addition to the selection on driver genes, the stabilizing selection wrought by chromosomal gene abundance. Further, we consider that the magnitude of selection pressure may not be a constant and implement a modifier to tune selection in our models. Lastly, we use our models as a quantitative measure of CIN that accounts for this selection.</p><p>Previous studies using single-cell sequencing identified surprisingly low karyotypic variance in human tumors including breast cancer (<xref ref-type="bibr" rid="bib27">Gao et al., 2016</xref>; <xref ref-type="bibr" rid="bib41">Kim et al., 2018</xref>; <xref ref-type="bibr" rid="bib76">Wang et al., 2014</xref>) and colorectal and ovarian cancer organoids (<xref ref-type="bibr" rid="bib8">Bolhaqueiro et al., 2019</xref>; <xref ref-type="bibr" rid="bib54">Nelson et al., 2020</xref>). It has been difficult to understand these findings in the light of widespread CIN in human cancer (<xref ref-type="bibr" rid="bib65">Sheltzer and Amon, 2011</xref>; <xref ref-type="bibr" rid="bib68">Silk et al., 2013</xref>; <xref ref-type="bibr" rid="bib75">Vasudevan et al., 2020</xref>; <xref ref-type="bibr" rid="bib78">Weaver et al., 2007</xref>; <xref ref-type="bibr" rid="bib80">Weaver and Cleveland, 2009</xref>). The best explanation of this apparent paradox is selection, which moderates karyotypic variance. Accounting for this, we can infer rates of chromosome mis-segregation in tumors or PDOs well within the range of rates observed microscopically in cancer cell lines. Additionally, no previous work, to our knowledge, has estimated the required sample size to infer CIN from scDNAseq data.</p><p>As described by others (<xref ref-type="bibr" rid="bib19">Dewhurst et al., 2014</xref>; <xref ref-type="bibr" rid="bib47">López et al., 2020</xref>), and consistent with our findings, early emergence of polyploid cells can markedly reduce apparent selection, leading to an elevated karyotype diversity over time. While we do not explicitly induce chance of whole genome doubling (WGD) events in simulations, populations that begin either diploid or tetraploid converge on near-triploid karyotypes over time, consistent with the notion that WGD can act as an evolutionary bridge to highly aneuploid karyotypes. Notably, our analysis indicates the samples with apparent polyploidy experienced among the lowest levels of karyotype selection.</p><p>In some early studies, CIN is considered a binary process—present or absent. We assumed that CIN measures are scalar, not binary, and measure this by rate of chromosome mis-segregation per division. A scalar is appropriate if, for example, there was a consistent probability of chromosome mis-segregation per division. However, we recognize that some mechanisms may not well adhere to this simplified model of CIN. For example, tumors with centrosome amplification may at times undergo bipolar division without mis-segregation, or, at other times, a multipolar division with extensive mis-segregation. Further, it is possible that some mechanisms may have correlated mis-segregations such that a daughter cell that gains one chromosome is more likely to gain other chromosomes, rather than lose them. Another possibility is that CIN could result in the mis-regulation of genes that further modify the rate of CIN. Our model does not yet account for punctuated behavior or changing rates of CIN. Furthermore, while recent studies have reported non-random mis-segregation of chromosomes (<xref ref-type="bibr" rid="bib20">Dumont et al., 2020</xref>; <xref ref-type="bibr" rid="bib82">Worrall et al., 2018</xref>), we did not incorporate these biases into our models as these studies do not reach consensus on which chromosomes are more frequently mis-segregated, which may be model-dependent.</p><p>Our approach reconstructs phylogenetic trees via copy number variation (CNV) analysis. This approach may be suboptimal given the selection on aneuploid states, and could be particularly problematic in the setting of convergent evolution. It is possible that this method results in low accuracy of the reconstructed phylogenies. Alternative approaches are possible, but would likely require re-design of the scDNAseq assay to include spiked-in primers that span highly polymorphic regions on each chromosome. If this were done, these sequences could be read in all cells and single-nucleotide polymorphisms could track individual maternal and paternal chromosomes, allowing a means of reconstructing cell phylogeny independent of CNVs. Despite this limitation, our phylogenetic reconstructions did seem to allow inference of CIN measures consistent with directly observed rates of chromosome mis-segregation in our taxol-induced CIN model as well as several independent cancer PDO models and cell lines.</p><p>A final limitation of our approach is we used previous estimates of cellular selection in our agent-based model and used these selection models to infer quantitative measures of CIN. While this approach seems to perform well in estimates of mis-segregation rates, we recognize that the selection models do not necessarily represent the real selective pressures on distinct aneuploidies. Future investigations are necessary to measure the selective pressure of distinct aneuploidies—a project that is now within technological reach. Selective pressures could also be influenced by cell type (<xref ref-type="bibr" rid="bib1">Auslander et al., 2019</xref>; <xref ref-type="bibr" rid="bib21">Dürrbaum et al., 2014</xref>; <xref ref-type="bibr" rid="bib61">Sack et al., 2018</xref>; <xref ref-type="bibr" rid="bib72">Starostik et al., 2020</xref>), tumor cell genetics (<xref ref-type="bibr" rid="bib24">Foijer et al., 2014</xref>; <xref ref-type="bibr" rid="bib30">Grim et al., 2012</xref>; <xref ref-type="bibr" rid="bib48">López-García et al., 2017</xref>; <xref ref-type="bibr" rid="bib69">Simões-Sousa et al., 2018</xref>; <xref ref-type="bibr" rid="bib70">Soto et al., 2017</xref>), and the microenvironment (<xref ref-type="bibr" rid="bib35">Hoevenaar et al., 2020</xref>).</p><p>In summation, we developed a theoretical and experimental framework for quantitative measure of chromosomal instability in human cancer. This framework accounts for selective pressure within tumors and employs Approximate Bayesian Computation, a commonly used analysis in evolutionary biology. Additionally, we determined that low-coverage single-cell DNA sequencing of at least 200 cells from a human tumor sample is sufficient to get an accurate ( > 90% accuracy) and reproducible measure of CIN. This work sets the stage for standardized quantitative measures of CIN that promise to clarify the underlying causes, consequences, and clinical utility of this nearly universal form of genomic instability.</p></sec><sec id="s4" sec-type="materials|methods"><title>Materials and methods</title><sec id="s4-1"><title>Agent-based modeling</title><p>Agent-based models were implemented using the agent-based platform, NetLogo 6.0.4 (<xref ref-type="bibr" rid="bib81">Wilensky, 1999</xref>).</p><sec id="s4-1-1"><title>Underlying assumptions for models of CIN and karyotype selection</title><p>Chromosome mis-segregation rate is defined as the number of chromosome missegregation events that occur per cellular division.</p><p>Cell division always results in 2 daughter cells.</p><p>Pmisseg,c is assigned uniformly for each cell in a population and for each chromosome.</p><p>Cells die when the copy number of any chromosome is equal to 0 or exceeding 6 unless otherwise noted.</p><p>Steps are based on the rate of division of euploid cells. We assume a probability of division (Pdivision) of 0.5, or half of the population divides every step, for euploid populations. This probabilistic division is to mimic the asynchrony of cellular proliferation and to allow for positive selection, where some cells may divide more rapidly than their euploid ancestors.</p><p>No chromosome is more likely to mis-segregate than any other.</p></sec></sec><sec id="s4-2"><title>Chromosome-arm scores</title><sec id="s4-2-1"><title>Gene abundance scores</title><p>The R package biomaRt v.2.46.3 was used to pull the chromosome arm location for each gene in Ensembl’s ‘Human genes’ dataset (GRCh38.p13). The number of genes on each chromosome arm were enumerated and Abundance scores were generated by normalizing the number of genes on each chromosome arm by the sum of all enumerated genes across chromosomes. Chromosome arms with no recorded genes were given a score of 0.</p></sec><sec id="s4-2-2"><title>Driver density scores</title><p>Arm-level ‘TSG-OG-Ess’ scores derived in <xref ref-type="bibr" rid="bib15">Davoli et al., 2013</xref> were adapted for our purposes. These values were derived from a pan-cancer analysis (TCGA) of the frequency of mutation of these genes and their location in the genome. These scores correlate with the frequency with which chromosomes are found to be amplified in the genome. We adapted these scores by normalizing the published ‘TSG-OG-Ess’ score for each chromosome arm by the sum of all Charm scores. Chromosome arms with no published Charm score were given a score of 0. We refer to these as TOE scores for our purposes.</p></sec><sec id="s4-2-3"><title>Hybrid scores</title><p>Chromosome arm scores for the Hybrid selection model are the average of the chromosome arm’s Gene Abundance and Driver Density scores.</p></sec></sec><sec id="s4-3"><title>Implementing karyotype selection</title><p>In each model, numerical scores are assigned to each chromosome, the sum of which represents the fitness of the karyotype (<xref ref-type="fig" rid="fig1">Figure 1B</xref>). At each simulation time step, fitness is re-calculated for each cell based on its updated karyotype. These fitness values determine if they undergo mitosis in the next round. However, the modality of selection changes how those karyotypes are assessed. Here, we implement four separate karyotype selection models (1) gene abundance, (2) driver density, (3) a hybrid gene abundance and driver density, and (4) neutral selection. The scores that are generated in each produce a fitness value (F) that can then be subjected to pressure (S) as described above.</p><sec id="s4-3-1"><title>Selection on gene abundance</title><p>The Gene Abundance selection model relies on the concept of gene dosage stoichiometry where the aneuploid karyotypes are selected against and that the extent of negative selection scales with the severity of aneuploidy and the identity and gene abundance on the aneuploid chromosomes (<xref ref-type="bibr" rid="bib65">Sheltzer and Amon, 2011</xref>). Chromosome arm fitness contribution scores (f<sub>c</sub>) are taken as the chromosome arm scores derived above (section 2.1) and the sum of these scores is 1. These base values are then modified under the gene abundance model to generate a contextual fitness score (CFS<sub>GA,c</sub>) at each time step such that…<disp-formula id="equ1"><mml:math id="m1"><mml:mrow><mml:msub><mml:mtext>CFS</mml:mtext><mml:mrow><mml:mtext>GA,c</mml:mtext></mml:mrow></mml:msub><mml:mtext>=</mml:mtext><mml:msub><mml:mtext>f</mml:mtext><mml:mrow><mml:mtext>c</mml:mtext></mml:mrow></mml:msub><mml:mtext>-</mml:mtext><mml:mfrac><mml:mrow><mml:msub><mml:mtext>f</mml:mtext><mml:mrow><mml:mtext>c</mml:mtext></mml:mrow></mml:msub><mml:mrow><mml:mo>×</mml:mo><mml:mrow><mml:mo stretchy="false">|</mml:mo></mml:mrow></mml:mrow><mml:msub><mml:mtext>n</mml:mtext><mml:mrow><mml:mtext>c</mml:mtext></mml:mrow></mml:msub><mml:mtext>-</mml:mtext><mml:msub><mml:mover><mml:mtext>x</mml:mtext><mml:mrow><mml:mo>−</mml:mo></mml:mrow></mml:mover><mml:mrow><mml:mtext>p</mml:mtext></mml:mrow></mml:msub><mml:mtext>|</mml:mtext></mml:mrow><mml:msub><mml:mover><mml:mtext>x</mml:mtext><mml:mrow><mml:mo>−</mml:mo></mml:mrow></mml:mover><mml:mrow><mml:mtext>p</mml:mtext></mml:mrow></mml:msub></mml:mfrac></mml:mrow></mml:math></disp-formula><disp-formula id="equ2"><mml:math id="m2"><mml:mrow><mml:mrow><mml:mi mathvariant="normal">F</mml:mi><mml:mo>=</mml:mo></mml:mrow><mml:munderover><mml:mo movablelimits="false">∑</mml:mo><mml:mrow><mml:mrow><mml:mi mathvariant="normal">c</mml:mi><mml:mo>=</mml:mo><mml:mn>1</mml:mn></mml:mrow></mml:mrow><mml:mrow><mml:mrow><mml:mn>46</mml:mn></mml:mrow></mml:mrow></mml:munderover><mml:msub><mml:mrow><mml:mi mathvariant="normal">C</mml:mi><mml:mi mathvariant="normal">F</mml:mi><mml:mi mathvariant="normal">S</mml:mi></mml:mrow><mml:mrow><mml:mrow><mml:mi mathvariant="normal">G</mml:mi><mml:mi mathvariant="normal">A</mml:mi><mml:mo>,</mml:mo><mml:mi mathvariant="normal">c</mml:mi></mml:mrow></mml:mrow></mml:msub></mml:mrow></mml:math></disp-formula></p><p>… where <inline-formula><mml:math id="inf2"><mml:mstyle displaystyle="true" scriptlevel="0"><mml:mrow><mml:msub><mml:mrow><mml:mover><mml:mi>X</mml:mi><mml:mo stretchy="false">¯</mml:mo></mml:mover></mml:mrow><mml:mrow><mml:mi>p</mml:mi></mml:mrow></mml:msub></mml:mrow></mml:mstyle></mml:math></inline-formula> is the average ploidy of the population and <inline-formula><mml:math id="inf3"><mml:mstyle displaystyle="true" scriptlevel="0"><mml:mrow><mml:msub><mml:mtext>n</mml:mtext><mml:mrow><mml:mtext>c</mml:mtext></mml:mrow></mml:msub></mml:mrow></mml:mstyle></mml:math></inline-formula> is the chromosome copy number. In this model, the fitness contribution of a chromosome declines as its distance from the average ploidy increases and that the magnitude of this effect is dependent on the size of the chromosome.</p></sec><sec id="s4-3-2"><title>Selection on driver density</title><p>The Driver Density modality relies on assigned fitness values to chromosomes based on their relative density of tumor suppressor genes, essential genes, and oncogenes. Chromosome arm fitness contribution scores (f<sub>c</sub>) are taken as the chromosome arm scores derived above (section Driver density scores) and are employed such that…<disp-formula id="equ3"><mml:math id="m3"><mml:mrow><mml:msub><mml:mtext>CFS</mml:mtext><mml:mrow><mml:mtext>TOE,c</mml:mtext></mml:mrow></mml:msub><mml:mtext>=</mml:mtext><mml:mfrac><mml:mrow><mml:msub><mml:mtext>n</mml:mtext><mml:mrow><mml:mtext>c</mml:mtext></mml:mrow></mml:msub><mml:mrow><mml:mo>×</mml:mo></mml:mrow><mml:msub><mml:mtext>TOE</mml:mtext><mml:mrow><mml:mtext>c</mml:mtext></mml:mrow></mml:msub></mml:mrow><mml:msub><mml:mover><mml:mtext>x</mml:mtext><mml:mrow><mml:mo>−</mml:mo></mml:mrow></mml:mover><mml:mrow><mml:mtext>p</mml:mtext></mml:mrow></mml:msub></mml:mfrac></mml:mrow></mml:math></disp-formula><disp-formula id="equ4"><mml:math id="m4"><mml:mrow><mml:mtext>F=</mml:mtext><mml:munderover><mml:mo movablelimits="false">∑</mml:mo><mml:mrow><mml:mtext>c=1</mml:mtext></mml:mrow><mml:mrow><mml:mtext>46</mml:mtext></mml:mrow></mml:munderover><mml:msub><mml:mtext>CFS</mml:mtext><mml:mrow><mml:mtext>TOE,c</mml:mtext></mml:mrow></mml:msub></mml:mrow></mml:math></disp-formula></p><p>This selection model benefits cells that have maximized the density of oncogenes and essential genes to tumor suppressors through chromosome mis-segregation.</p></sec><sec id="s4-3-3"><title>Hybrid selection</title><p>The hybrid model relies on selection on both gene abundance and driver densities. CFS<sub>TOE,c</sub> and CFS<sub>GA,c</sub> are both calculated and averaged such that…<disp-formula id="equ5"><mml:math id="m5"><mml:mtext>F = </mml:mtext><mml:mrow><mml:munderover><mml:mo>∑</mml:mo><mml:mrow><mml:mtext>c=1</mml:mtext></mml:mrow><mml:mrow><mml:mtext>46</mml:mtext></mml:mrow></mml:munderover><mml:mrow><mml:mfrac><mml:mrow><mml:msub><mml:mrow><mml:mtext>CFS</mml:mtext></mml:mrow><mml:mrow><mml:mtext>GA,c</mml:mtext></mml:mrow></mml:msub><mml:mo>+</mml:mo><mml:msub><mml:mrow><mml:mtext>CFS</mml:mtext></mml:mrow><mml:mrow><mml:mtext>TOE,c</mml:mtext></mml:mrow></mml:msub></mml:mrow><mml:mrow><mml:mtext>2</mml:mtext></mml:mrow></mml:mfrac></mml:mrow></mml:mrow></mml:math></disp-formula></p></sec><sec id="s4-3-4"><title>Neutral selection</title><p>When populations are grown under neutral selection, the fitness of each cell is constitutively set to 1 regardless of the cells’ individual karyotypes.<disp-formula id="equ6"><mml:math id="m6"><mml:mtext>F = 1</mml:mtext></mml:math></disp-formula></p></sec><sec id="s4-3-5"><title>Scaling selection pressure</title><p>Within each model of karyotype selection, the magnitude of selective pressure upon any karyotype, with fitness F, can be scaled by applying the scalar exponent S to produce a modified fitness score F<sub>M</sub>. Thus…<disp-formula id="equ7"><mml:math id="m7"><mml:msub><mml:mrow><mml:mtext>F</mml:mtext></mml:mrow><mml:mrow><mml:mtext>M</mml:mtext></mml:mrow></mml:msub><mml:mtext> = </mml:mtext><mml:msup><mml:mrow><mml:mtext>F</mml:mtext></mml:mrow><mml:mrow><mml:mtext>S</mml:mtext></mml:mrow></mml:msup></mml:math></disp-formula></p><p>For example, in the Gene Abundance model of karyotype selection, an otherwise diploid cell with three copies of chromosome 1 in a diploid population will have a F value of 0.954. Under selection-null conditions (S = 0)…<disp-formula id="equ8"><mml:math id="m8"><mml:mrow><mml:msub><mml:mtext>F</mml:mtext><mml:mrow><mml:mtext>M</mml:mtext></mml:mrow></mml:msub><mml:mtext>=</mml:mtext><mml:msup><mml:mtext>F</mml:mtext><mml:mrow><mml:mtext>S</mml:mtext></mml:mrow></mml:msup><mml:mtext>=</mml:mtext><mml:msup><mml:mtext>0.954</mml:mtext><mml:mrow><mml:mtext>0</mml:mtext></mml:mrow></mml:msup><mml:mtext>=</mml:mtext><mml:mn>1</mml:mn></mml:mrow></mml:math></disp-formula></p><p>… the fitness of the aneuploid cell is equivalent to that of a euploid cell. Under conditions of high selection (S = 50)…<disp-formula id="equ9"><mml:math id="m9"><mml:mrow><mml:msub><mml:mtext>F</mml:mtext><mml:mrow><mml:mtext>M</mml:mtext></mml:mrow></mml:msub><mml:mtext>=</mml:mtext><mml:msup><mml:mtext>F</mml:mtext><mml:mrow><mml:mtext>S</mml:mtext></mml:mrow></mml:msup><mml:mtext>=</mml:mtext><mml:msup><mml:mtext>0.954</mml:mtext><mml:mrow><mml:mtext>50</mml:mtext></mml:mrow></mml:msup><mml:mtext>=</mml:mtext><mml:mn>0.097</mml:mn></mml:mrow></mml:math></disp-formula></p><p>…fitness of the aneuploid cell is ~10% that of the euploid cell and thus divides ~10% as frequently.</p></sec></sec><sec id="s4-4"><title>Modeling growing and constant population dynamics</title><p>To accommodate different population size dynamics, we implemented our model using either growing, pseudo-Moran limited population dynamics and constant-size populations with approximated Wright-Fisher population dynamics.</p><sec id="s4-4-1"><title>Simulating CIN in exponentially growing populations with pseudo-Moran limits</title><p>Populations begin with 100 founder cells with a euploid karyotype of integer value <inline-formula><mml:math id="inf4"><mml:mstyle displaystyle="true" scriptlevel="0"><mml:mrow><mml:msub><mml:mrow><mml:mover><mml:mi>X</mml:mi><mml:mo stretchy="false">¯</mml:mo></mml:mover></mml:mrow><mml:mi>p</mml:mi></mml:msub></mml:mrow></mml:mstyle></mml:math></inline-formula> and the simulation is initiated.</p><p>CFS values are calculated for each chromosome in a cell according to the chosen karyotype selection model.</p><p>Cellular fitness is calculated based on CFS values.</p><p>Selective pressure (S) is applied to fitness (F) values to modify cellular fitness (F<sub>M</sub>).</p><p>Cells are checked to see if any death conditions are met and if the population limit is met. Cells die if any chromosome arm copy (n<sub>c</sub>) is less than 1 or greater than 6 (unless otherwise indicated). We implemented population size limits in a pseudo-Moran fashion to reduce computational constraints. If the population size is 3000 cells or greater, a random half of the population is deleted.</p><p>Cells probabilistically divide if their fitness is greater than a random float (R) between 0 and 2. Thus...<disp-formula id="equ10"><mml:math id="m10"><mml:mrow><mml:mi mathvariant="normal">R</mml:mi><mml:mo>∼</mml:mo><mml:mi mathvariant="normal">U</mml:mi><mml:mo stretchy="false">[</mml:mo><mml:mn>0</mml:mn><mml:mo>,</mml:mo><mml:mspace width="thinmathspace"/><mml:mn>1</mml:mn><mml:mo stretchy="false">]</mml:mo></mml:mrow></mml:math></disp-formula></p><p>If a cell does not divide, it restarts the cycle from CFS values are calculated for each chromosome in a cell according to the chosen karyotype selection model. If a cell divides, mis-segregations may occur.</p><p>Each copy (n<sub>c</sub>) of each chromosome (c) has an opportunity to mis-segregate probabilistically. For each chromosome copy, a mis-segregation occurs if a random float (R) between 0–1 falls below P<sub>misseg</sub>. Thus...<disp-formula id="equ11"><mml:math id="m11"><mml:mrow><mml:mi mathvariant="normal">R</mml:mi><mml:mo>∼</mml:mo><mml:mi mathvariant="normal">U</mml:mi><mml:mo stretchy="false">[</mml:mo><mml:mn>0</mml:mn><mml:mo>,</mml:mo><mml:mspace width="thinmathspace"/><mml:mn>1</mml:mn><mml:mo stretchy="false">]</mml:mo></mml:mrow></mml:math></disp-formula><disp-formula id="equ12"><mml:math id="m12"><mml:mrow><mml:mi mathvariant="normal">M</mml:mi><mml:mi mathvariant="normal">i</mml:mi><mml:mi mathvariant="normal">s</mml:mi><mml:mo>−</mml:mo><mml:mi mathvariant="normal">s</mml:mi><mml:mi mathvariant="normal">e</mml:mi><mml:mi mathvariant="normal">g</mml:mi><mml:mi mathvariant="normal">r</mml:mi><mml:mi mathvariant="normal">e</mml:mi><mml:mi mathvariant="normal">g</mml:mi><mml:mi mathvariant="normal">a</mml:mi><mml:mi mathvariant="normal">t</mml:mi><mml:mi mathvariant="normal">e</mml:mi><mml:mspace width="thinmathspace"/><mml:mi mathvariant="normal">c</mml:mi><mml:mi mathvariant="normal">h</mml:mi><mml:mi mathvariant="normal">r</mml:mi><mml:mi mathvariant="normal">o</mml:mi><mml:mi mathvariant="normal">m</mml:mi><mml:mi mathvariant="normal">o</mml:mi><mml:mi mathvariant="normal">s</mml:mi><mml:mi mathvariant="normal">o</mml:mi><mml:mi mathvariant="normal">m</mml:mi><mml:mi mathvariant="normal">e</mml:mi><mml:mspace width="thinmathspace"/><mml:mi mathvariant="normal">c</mml:mi><mml:mspace width="thinmathspace"/><mml:mi mathvariant="normal">i</mml:mi><mml:mi mathvariant="normal">f</mml:mi><mml:mspace width="thinmathspace"/><mml:msub><mml:mtext>P</mml:mtext><mml:mrow><mml:mi mathvariant="normal">m</mml:mi><mml:mi mathvariant="normal">i</mml:mi><mml:mi mathvariant="normal">s</mml:mi><mml:mi mathvariant="normal">s</mml:mi><mml:mi mathvariant="normal">e</mml:mi><mml:mi mathvariant="normal">g</mml:mi><mml:mo>,</mml:mo><mml:mi mathvariant="normal">c</mml:mi><mml:mo>></mml:mo><mml:mi mathvariant="normal">R</mml:mi></mml:mrow></mml:msub></mml:mrow></mml:math></disp-formula></p><p>If a chromosome copy is not mis-segregated, the next chromosome copy is tested. If a chromosome copy is mis-segregated, chromosome arms may be segregated separately (i.e. a reciprocal, arm-level CNA) if a random float (R) between 0 and 1 falls below P<sub>break</sub>. Thus...<disp-formula id="equ13"><mml:math id="m13"><mml:mrow><mml:mi mathvariant="normal">R</mml:mi><mml:mo>∼</mml:mo><mml:mi mathvariant="normal">U</mml:mi><mml:mo stretchy="false">[</mml:mo><mml:mn>0</mml:mn><mml:mo>,</mml:mo><mml:mspace width="thinmathspace"/><mml:mn>1</mml:mn><mml:mo stretchy="false">]</mml:mo></mml:mrow></mml:math></disp-formula><disp-formula id="equ14"><mml:math id="m14"><mml:mrow><mml:mi mathvariant="normal">B</mml:mi><mml:mi mathvariant="normal">r</mml:mi><mml:mi mathvariant="normal">e</mml:mi><mml:mi mathvariant="normal">a</mml:mi><mml:mi mathvariant="normal">k</mml:mi><mml:mspace width="thinmathspace"/><mml:mi mathvariant="normal">c</mml:mi><mml:mi mathvariant="normal">h</mml:mi><mml:mi mathvariant="normal">r</mml:mi><mml:mi mathvariant="normal">o</mml:mi><mml:mi mathvariant="normal">m</mml:mi><mml:mi mathvariant="normal">o</mml:mi><mml:mi mathvariant="normal">s</mml:mi><mml:mi mathvariant="normal">o</mml:mi><mml:mi mathvariant="normal">m</mml:mi><mml:mi mathvariant="normal">e</mml:mi><mml:mspace width="thinmathspace"/><mml:mi mathvariant="normal">c</mml:mi><mml:mspace width="thinmathspace"/><mml:mi mathvariant="normal">i</mml:mi><mml:mi mathvariant="normal">f</mml:mi><mml:mspace width="thinmathspace"/><mml:msub><mml:mrow><mml:mi mathvariant="normal">P</mml:mi></mml:mrow><mml:mrow><mml:mi mathvariant="normal">m</mml:mi><mml:mi mathvariant="normal">i</mml:mi><mml:mi mathvariant="normal">s</mml:mi><mml:mi mathvariant="normal">s</mml:mi><mml:mi mathvariant="normal">e</mml:mi><mml:mi mathvariant="normal">g</mml:mi><mml:mo>,</mml:mo><mml:mi mathvariant="normal">c</mml:mi><mml:mo>></mml:mo><mml:mi mathvariant="normal">R</mml:mi></mml:mrow></mml:msub></mml:mrow></mml:math></disp-formula></p><p>The karyotype of the cell is modified according to the results of the mis-segregation sequence above. When the mis-segregation sequence is complete, a clone of the initial cell with any reciprocal copy number alterations to its karyotype is created.</p><p>The simulation ends if it reaches 100 steps and data are exported. Otherwise, the simulation continues from CFS values are calculated for each chromosome in a cell according to the chosen karyotype selection model.</p></sec><sec id="s4-4-2"><title>Simulating CIN in constant-size populations with approximated Wright-Fisher dynamics</title><p>We approximated constant-size Wright-Fisher dynamics in our model by re-initiating the population at each time step and randomly drawing from the previous generation’s distribution of chromosome copy numbers for each chromosome in each cell of the new population. Because the exponential pseudo-Moran model relies on proliferation rates across over-lapping generations to enact karyotype selection, such a method would not be useful here. To accommodate karyotype selection in this model, we employed an additional baseline death rate of about 20% (<xref ref-type="bibr" rid="bib71">Sottoriva et al., 2015</xref>) that increases for cells with lower fitness and decreases for cells with higher fitness (see section 4.2.9). In this way, the karyotypes of the cells that die are removed from the pool of karyotypes that are drawn upon in the subsequent generation. CIN is simulated in this model as follows:</p><p>Populations begin with 4,500 founder cells and the simulation is (re-)initiated. The population begins with a euploid karyotype of integer value <inline-formula><mml:math id="inf5"><mml:mstyle displaystyle="true" scriptlevel="0"><mml:mrow><mml:msub><mml:mrow><mml:mover><mml:mi>X</mml:mi><mml:mo stretchy="false">¯</mml:mo></mml:mover></mml:mrow><mml:mrow><mml:mrow><mml:mi>p</mml:mi></mml:mrow></mml:mrow></mml:msub></mml:mrow></mml:mstyle></mml:math></inline-formula> if the population is being created for the first time.</p><p>Cells divide every step, regardless of fitness.</p><p>Chromosomes are mis-segregated in the same fashion as the exponential pseudo-Moran model above (sections 4.1.8–4.1.10).</p><p>The simulation ends if it reaches 100 steps and data are exported. Otherwise, the simulation continues from 4.2.1.</p><p>CFS values are calculated for each chromosome in a cell according to the chosen karyotype selection model.</p><p>Cellular fitness is calculated based on CFS values.</p><p>Selective pressure (S) is applied to fitness (F) values to modify cellular fitness (F<sub>M</sub>).</p><p>Cells are checked to see if any death conditions are met and if the population limit is met. Cells die if any chromosome arm copy (n<sub>c</sub>) is less than 1 or greater than 6 (unless otherwise indicated).</p><p>Additionally, the cells’ fitness values and a random float (R) between 0 and 5 are used to determine if they die. In this way, a cell with a fitness of 1 has a 20% baseline death rate. Thus, cells die if…<disp-formula id="equ15"><mml:math id="m15"><mml:mrow><mml:mfrac><mml:mtext>1</mml:mtext><mml:mrow><mml:msup><mml:mrow><mml:mi mathvariant="normal">F</mml:mi></mml:mrow><mml:mrow><mml:mrow><mml:mi mathvariant="normal">S</mml:mi></mml:mrow></mml:mrow></mml:msup><mml:mrow><mml:mo>+</mml:mo><mml:mn>0.001</mml:mn></mml:mrow></mml:mrow></mml:mfrac><mml:mrow><mml:mo>></mml:mo><mml:mi mathvariant="normal">R</mml:mi><mml:mo>∼</mml:mo><mml:mi mathvariant="normal">U</mml:mi><mml:mo stretchy="false">[</mml:mo><mml:mn>0</mml:mn><mml:mo>,</mml:mo><mml:mn>5</mml:mn><mml:mo stretchy="false">]</mml:mo></mml:mrow></mml:mrow></mml:math></disp-formula></p><p>After determining cell death, the copy number distributions of each cells’ chromosome arm (c) are individually stored.</p><p>The cycle repeats from 4.2.1. However, the re-initated population will have its chromosome arm copy numbers drawn from the previous generation’s stored chromosome arm copy number distributions.</p></sec></sec><sec id="s4-5"><title>Analysis of population diversity and topology in biological and simulated data</title><p>Phylogenetic trees were reconstructed from chromosome copy number profiles from live and simulated cells by calculating pairwise Euclidean distance matrices and performing complete-linkage clustering in R (<xref ref-type="bibr" rid="bib59">R Development Core Team, 2021</xref>). Phylogenetic tree topology measurements were performed in R using the package phyloTop v2.1.1 (<xref ref-type="bibr" rid="bib40">Kendall et al., 2018</xref>). Sackin and Colless indices of tree imbalance were calculated, normalizing to the number of tree tips. Cherry and pitchfork number were also normalized to the size of the tree. MKV is taken as the variance of individual chromosomes taken across the population, averaged across all chromosomes, then normalized to the average ploidy of the population. Average aneuploidy is calculated as the variance within a single cell’s karyotype averaged across the population.</p></sec><sec id="s4-6"><title>Approximate bayesian computation</title><p>Approximate Bayesian computation was used for parameter inference of experimental data from simulated data. For this we employed the the “abc” function in the R package abc v2.1 (<xref ref-type="bibr" rid="bib13">Csilléry et al., 2010</xref>). In short, a set of simulation parameters, <italic>θ<sub>i</sub></italic>, is sampled from the prior distribution. This set of parameters corresponds to a set of simulated summary statistics, <italic>S(y<sub>i</sub></italic>), in this case phylogenetic tree shapes, which can be compared to the set of experimental summary statistics, <italic>S(y<sub>o</sub></italic>). The Euclidean distance between the experimental and simulated summary statistics can then be calculated (<italic>dS(y<sub>i</sub>),S(y<sub>o</sub></italic>)). A threshold, <italic>T,</italic> is then selected—0.05 in our case—which rejects the lower 1 <italic>T</italic> sets of simulation parameters that correspond. The remaining parameters represent those that gave summary statistics with the highest similarity to the experimental summary statistics. These represent the posterior distribution of accepted parameters.</p><p>Bayesian model selection was performed using the “postpr” function in the same R package using tolerance threshold of 0.05 and rejection sampling method. This was used to calculate the posterior probability of each selection model <italic>within</italic> each growth model and the Bayes factor for each selection model with neutral selection as the null hypothesis. Bayes factors > 5 were considered substantial evidence of the alternative hypothesis.</p></sec><sec id="s4-7"><title>Sliding window analysis to tune time-steps for approximate Bayesian computation</title><p>We chose which simulation time steps to use for approximate Bayesian computation on organoid and biopsy data by repeating the inference using a sliding window of prior datasets with a width of 11 time steps (i.e. parameters from steps ∈ [0–10], [10-20], …, [91-100]) to see if the posterior distributions would stabilize over time. We then chose simulations from 40 to 80 time steps as our prior dataset as this range provided both a stable inference and is centered around 60 time steps (analogous to 30 generations, estimated to generate a 1 cm palpable mass of ~1 billion cells).</p></sec><sec id="s4-8"><title>Cell cultivation procedures</title><p>Cal51 cells expressing stably integrated RFP-tagged histone H2B and GFP-tagged a-tubulin were generated as previously described (<xref ref-type="bibr" rid="bib83">Zasadil et al., 2014</xref>). Cells were maintained at 37 ºC and 5% CO<sub>2</sub> in a humidified, water-jacketed incubator and propagated in Dulbecco’s Modified Eagle’s Medium (DMEM) – High Glucose formulation (Cat #: 11965118) supplemented with 10% fetal bovine serum and 100 units/mL penicillin-streptomycin. Paclitaxel (Tocris Bioscience, Cat #: 1097/10) used for cell culture experiments was dissolved in DMSO. The Cal51 cells were obtained from the DSMZ-German Collection of Microorganisms and Cell Cultures and were free from mycoplasma contamination prior to study. Karyotype analysis confirms the near-diploid characteristic of the cell line and the presence of both fluorescent markers suggests they are free of other contaminating cell lines.</p></sec><sec id="s4-9"><title>Time-lapse fluorescence microscopy</title><p>Cal51 cells were transduced with lentivirus expressing mNeonGreen-tubulin-P2A-H2B-FusionRed. A monoclonal line was treated with 20 nM paclitaxel for 24, 48, or 72 hr before timelapse analysis at 37 <sup>o</sup>C and 10% CO<sub>2</sub>. Five 2 µm z-plane images were acquired using a Nikon Ti-E inverted microscope with a cMos camera at 3-min intervals using a 40 X/0.75 NA objective lens and Nikon Elements software.</p></sec><sec id="s4-10"><title>Flow cytometric analysis and cell sorting</title><p>Cells were harvested with trypsin, passed through a 35 μm mesh filter, and rinsed with PBS prior to fixation in ice cold 80% methanol. Fixed cells were stored at –80 ºC until analysis and sorting at which point fixed cells were resuspended in PBS containing 10 μg/ml DAPI for cell cycle analysis.</p><sec id="s4-10-1"><title>Flow cytometric analysis</title><p>Initial DNA content and cell cycle analyses were performed on a 5 laser BD LSR II. Doublets were excluded from analysis via standard FSC/SSC gating procedures. DNA content was analyzed via DAPI excitation at 355 nm and 450/50 emission using a 410 nm long pass dichroic filter.</p></sec><sec id="s4-10-2"><title>Fluorescence activated cell sorting</title><p>Cell sorting was performed using the same analysis procedures described above on a BD FACS AriaII cell sorter. In general, single cells were sorted through a 130 μm low-pressure deposition nozzle into each well of a 96-well PCR plate containing 10 μl Lysis and Fragmentation Buffer cooled to 4 ºC on a Eppendorf PCR plate cooler. Immediately after sorting PCR plates were centrifuged at 300 x g for 60 s. For comparison of single-cell sequencing to bulk sequencing, 1000 cells were sorted into each ‘bulk’ well. The index of sorted cells was retained allowing for the post hoc estimation of DNA content for each cell.</p></sec></sec><sec id="s4-11"><title>Low-coverage single-cell whole genome sequencing</title><p>Initial library preparation for low-coverage scDNAseq was performed as previously described (<xref ref-type="bibr" rid="bib45">Leung et al., 2016</xref>) and adapted for low coverage whole genome sequencing instead of high coverage targeted sequencing. Initial genome amplification was performed using the GenomePlex Single Cell Whole Genome Amplification Kit and protocol (Sigma Aldrich, Cat #: WGA4). Cells were sorted into 10 μl pre-prepared Lysis and Fragmentation buffer containing Proteinase K. DNA was fragmented to an average of 1 kb in length prior to amplification. Single cell libraries were purified on a 96-well column plate (Promega, Cat #: A2271). Library fragment distribution was assessed via agarose gel electrophoresis and concentrations were measured on a Nanodrop 2000. Sequencing libraries were prepared using the QuantaBio sparQ DNA Frag and Library Prep Kit. Amplified single-cell DNA was enzymatically fragmented to ~250 bp, 5’-phosphorylated, and 3’-dA-tailed. Custom Illumina adapters with 96 unique 8 bp P7 index barcodes were ligated to individual libraries to enable multiplexed sequencing (<xref ref-type="bibr" rid="bib45">Leung et al., 2016</xref>). Barcoded libraries were amplified following size selection via AxygenAxyPrep Mag beads (Cat #: 14-223-152). Amplified library DNA concentration was quantified using the Quant-iT Broad-Range dsDNA Assay Kit (Thermo, Cat #: Q33130). Single-cell libraries were pooled to 15 nM and final concentration was measured via qPCR. Single-end 100 bp sequencing was performed on an Illumina HiSeq2500.</p></sec><sec id="s4-12"><title>Single-cell copy number sequencing data processing</title><p>Single-cell DNA sequence reads were demultiplexed using unique barcode index sequences and trimmed to remove adapter sequences. Reads were aligned to GRCh38 using Bowtie2. Aligned BAM files were then processed using Ginkgo to make binned copy number calls. Reads are aligned within 500 kb bins and estimated DNA content for each cell, obtained by flow cytometric analysis, was used to calculate bin copy numbers based on the relative ratio of reads per bin (<xref ref-type="bibr" rid="bib28">Garvin et al., 2015</xref>). We modified and ran Ginkgo locally to allow for the analysis of highly variable karyotypes with low ploidy values (see Code and Data Availability). Whole-chromosome copy number calls were calculated as the modal binned copy number across an individual chromosome. Cells with fewer than 100,000 reads were filtered out to ensure accurate copy number calls (<xref ref-type="bibr" rid="bib6">Baslan et al., 2015</xref>). Cells whose predicted ploidy deviated more than 32% from the observed ploidy by FACS were also filtered out. The final coverage for the filtered dataset was 0.03 (5). Single cell data extracted from <xref ref-type="bibr" rid="bib52">Navin et al., 2011</xref> were separated into their individual clones and depleted of euploid cells. Single cell data from <xref ref-type="bibr" rid="bib8">Bolhaqueiro et al., 2019</xref> were filtered to include only the aneuploid data that fell within the ploidies observed in the study (see Code and Data Availability).</p></sec><sec id="s4-13"><title>Review and approximation of mis-segregation rates from published Studies</title><p>We reviewed the literature to extract per chromosome rates of mis-segregation for cell lines and clinical samples. Some studies publish these rates. For those that did not, we estimated these rates by approximating the plotted incidence of segregation errors thusly:<disp-formula id="equ16"><mml:math id="m16"><mml:mrow><mml:mi mathvariant="normal">A</mml:mi><mml:mi mathvariant="normal">p</mml:mi><mml:mi mathvariant="normal">p</mml:mi><mml:mi mathvariant="normal">r</mml:mi><mml:mi mathvariant="normal">o</mml:mi><mml:mi mathvariant="normal">x</mml:mi><mml:mi mathvariant="normal">i</mml:mi><mml:mi mathvariant="normal">m</mml:mi><mml:mi mathvariant="normal">a</mml:mi><mml:mi mathvariant="normal">t</mml:mi><mml:mi mathvariant="normal">e</mml:mi><mml:mspace width="thinmathspace"/><mml:mi mathvariant="normal">m</mml:mi><mml:mi mathvariant="normal">i</mml:mi><mml:mi mathvariant="normal">s</mml:mi><mml:mi mathvariant="normal">s</mml:mi><mml:mi mathvariant="normal">e</mml:mi><mml:mi mathvariant="normal">g</mml:mi><mml:mi mathvariant="normal">r</mml:mi><mml:mi mathvariant="normal">e</mml:mi><mml:mi mathvariant="normal">g</mml:mi><mml:mi mathvariant="normal">r</mml:mi><mml:mi mathvariant="normal">a</mml:mi><mml:mi mathvariant="normal">t</mml:mi><mml:mi mathvariant="normal">i</mml:mi><mml:mi mathvariant="normal">o</mml:mi><mml:mi mathvariant="normal">n</mml:mi><mml:mspace width="thinmathspace"/><mml:mi mathvariant="normal">r</mml:mi><mml:mi mathvariant="normal">a</mml:mi><mml:mi mathvariant="normal">t</mml:mi><mml:mi mathvariant="normal">e</mml:mi><mml:mspace width="thinmathspace"/><mml:mi mathvariant="normal">p</mml:mi><mml:mi mathvariant="normal">e</mml:mi><mml:mi mathvariant="normal">r</mml:mi><mml:mspace width="thinmathspace"/><mml:mi mathvariant="normal">c</mml:mi><mml:mi mathvariant="normal">h</mml:mi><mml:mi mathvariant="normal">r</mml:mi><mml:mi mathvariant="normal">o</mml:mi><mml:mi mathvariant="normal">m</mml:mi><mml:mi mathvariant="normal">o</mml:mi><mml:mi mathvariant="normal">s</mml:mi><mml:mi mathvariant="normal">o</mml:mi><mml:mi mathvariant="normal">m</mml:mi><mml:mi mathvariant="normal">e</mml:mi><mml:mo>=</mml:mo><mml:mfrac><mml:mrow><mml:mi mathvariant="normal">O</mml:mi><mml:mi mathvariant="normal">b</mml:mi><mml:mi mathvariant="normal">s</mml:mi><mml:mi mathvariant="normal">e</mml:mi><mml:mi mathvariant="normal">r</mml:mi><mml:mi mathvariant="normal">v</mml:mi><mml:mi mathvariant="normal">e</mml:mi><mml:mi mathvariant="normal">d</mml:mi><mml:mspace width="thinmathspace"/><mml:mi mathvariant="normal">%</mml:mi><mml:mspace width="thinmathspace"/><mml:mi mathvariant="normal">f</mml:mi><mml:mi mathvariant="normal">r</mml:mi><mml:mi mathvariant="normal">e</mml:mi><mml:mi mathvariant="normal">q</mml:mi><mml:mi mathvariant="normal">u</mml:mi><mml:mi mathvariant="normal">e</mml:mi><mml:mi mathvariant="normal">n</mml:mi><mml:mi mathvariant="normal">c</mml:mi><mml:mi mathvariant="normal">y</mml:mi><mml:mspace width="thinmathspace"/><mml:mi mathvariant="normal">o</mml:mi><mml:mi mathvariant="normal">f</mml:mi><mml:mspace width="thinmathspace"/><mml:mi mathvariant="normal">e</mml:mi><mml:mi mathvariant="normal">r</mml:mi><mml:mi mathvariant="normal">r</mml:mi><mml:mi mathvariant="normal">o</mml:mi><mml:mi mathvariant="normal">r</mml:mi><mml:mi mathvariant="normal">s</mml:mi><mml:mspace width="thinmathspace"/><mml:mi mathvariant="normal">p</mml:mi><mml:mi mathvariant="normal">e</mml:mi><mml:mi mathvariant="normal">r</mml:mi><mml:mspace width="thinmathspace"/><mml:mi mathvariant="normal">d</mml:mi><mml:mi mathvariant="normal">i</mml:mi><mml:mi mathvariant="normal">v</mml:mi><mml:mi mathvariant="normal">i</mml:mi><mml:mi mathvariant="normal">s</mml:mi><mml:mi mathvariant="normal">i</mml:mi><mml:mi mathvariant="normal">o</mml:mi><mml:mi mathvariant="normal">n</mml:mi><mml:mrow><mml:mo>/</mml:mo></mml:mrow><mml:mn>100</mml:mn></mml:mrow><mml:mrow><mml:mi mathvariant="normal">T</mml:mi><mml:mi mathvariant="normal">o</mml:mi><mml:mi mathvariant="normal">t</mml:mi><mml:mi mathvariant="normal">a</mml:mi><mml:mi mathvariant="normal">l</mml:mi><mml:mspace width="thinmathspace"/><mml:mi mathvariant="normal">#</mml:mi><mml:mspace width="thinmathspace"/><mml:mi mathvariant="normal">m</mml:mi><mml:mi mathvariant="normal">o</mml:mi><mml:mi mathvariant="normal">d</mml:mi><mml:mi mathvariant="normal">a</mml:mi><mml:mi mathvariant="normal">l</mml:mi><mml:mspace width="thinmathspace"/><mml:mi mathvariant="normal">c</mml:mi><mml:mi mathvariant="normal">h</mml:mi><mml:mi mathvariant="normal">r</mml:mi><mml:mi mathvariant="normal">o</mml:mi><mml:mi mathvariant="normal">m</mml:mi><mml:mi mathvariant="normal">o</mml:mi><mml:mi mathvariant="normal">s</mml:mi><mml:mi mathvariant="normal">o</mml:mi><mml:mi mathvariant="normal">m</mml:mi><mml:mi mathvariant="normal">e</mml:mi><mml:mi mathvariant="normal">s</mml:mi><mml:mspace width="thinmathspace"/><mml:mi mathvariant="normal">i</mml:mi><mml:mi mathvariant="normal">n</mml:mi><mml:mspace width="thinmathspace"/><mml:mi mathvariant="normal">s</mml:mi><mml:mi mathvariant="normal">a</mml:mi><mml:mi mathvariant="normal">m</mml:mi><mml:mi mathvariant="normal">p</mml:mi><mml:mi mathvariant="normal">l</mml:mi><mml:mi mathvariant="normal">e</mml:mi></mml:mrow></mml:mfrac></mml:mrow></mml:math></disp-formula></p><p>Modal chromosome numbers were either taken from ATCC where available or were assumed to equal 46. Observed % frequencies were approximated from published plots. Approximated rates assume that 1 chromosome is mis-segregated at a time.</p></sec></sec></body><back><sec id="s5" sec-type="additional-information"><title>Additional information</title><fn-group content-type="competing-interest"><title>Competing interests</title><fn fn-type="COI-statement" id="conf1"><p>No competing interests declared</p></fn><fn fn-type="COI-statement" id="conf2"><p>No competing interests declared</p></fn><fn fn-type="COI-statement" id="conf3"><p>declares the following: Medical advisory board of Strata Oncology; Research funding from Abbvie, Genentech, Puma, Arcus, Apollomics, Loxo Oncology/Lilly, and Elevation Oncology. I hold patents on microfluidic device for drug testing, and for homologous recombination and super-resolution microscopy technologies. I declare all interests without adjudicating relationship to the published work</p></fn></fn-group><fn-group content-type="author-contribution"><title>Author contributions</title><fn fn-type="con" id="con1"><p>Conceptualization, Data curation, Formal analysis, Investigation, Methodology, Software</p></fn><fn fn-type="con" id="con2"><p>Formal analysis, Investigation, Writing – review and editing</p></fn><fn fn-type="con" id="con3"><p>Investigation, Writing – review and editing</p></fn><fn fn-type="con" id="con4"><p>Funding acquisition, Project administration, Supervision, Writing – review and editing</p></fn><fn fn-type="con" id="con5"><p>Conceptualization, Funding acquisition, Methodology, Project administration, Resources, Supervision, Writing – original draft, Writing – review and editing</p></fn></fn-group></sec><sec id="s6" sec-type="supplementary-material"><title>Additional files</title><supplementary-material id="transrepform"><label>Transparent reporting form</label><media mime-subtype="docx" mimetype="application" xlink:href="elife-69799-transrepform1-v2.docx"/></supplementary-material></sec><sec id="s7" sec-type="data-availability"><title>Data availability</title><p>Single-cell DNA sequencing data from this study has been deposited in NCBI SRA (PRJNA725515). All data and scripts used for modeling and analysis have been deposited in OSF at <ext-link ext-link-type="uri" xlink:href="https://osf.io/snrg3/">https://osf.io/snrg3/</ext-link>.</p><p>The following datasets were generated:</p><p><element-citation id="dataset1" publication-type="data" specific-use="isSupplementedBy"><person-group person-group-type="author"><name><surname>Lynch</surname><given-names>AR</given-names></name><name><surname>Arp</surname><given-names>NL</given-names></name><name><surname>Zhou</surname><given-names>AS</given-names></name><name><surname>Weaver</surname><given-names>BA</given-names></name><name><surname>Burkard</surname><given-names>ME</given-names></name></person-group><year iso-8601-date="2021">2021</year><data-title>Quantifying chromosomal instability from intratumoral karyotype diversity using agent- based modeling and Bayesian inference</data-title><source>Open Science Framework</source><pub-id pub-id-type="doi">10.17605/OSF.IO/SNRG3</pub-id></element-citation></p><p><element-citation id="dataset2" publication-type="data" specific-use="isSupplementedBy"><person-group person-group-type="author"><name><surname>Lynch</surname><given-names>AR</given-names></name><name><surname>Arp</surname><given-names>NL</given-names></name><name><surname>Zhou</surname><given-names>AS</given-names></name><name><surname>Weaver</surname><given-names>BA</given-names></name><name><surname>Burkard</surname><given-names>ME</given-names></name></person-group><year iso-8601-date="2021">2021</year><data-title>Quantifying chromosomal instability from intratumoral karyotype diversity Quantifying chromosomal instability from intratumoral karyotype diversity</data-title><source>NCBI BioProject</source><pub-id pub-id-type="accession" xlink:href="https://www.ncbi.nlm.nih.gov/bioproject/PRJNA725515">PRJNA725515</pub-id></element-citation></p></sec><ack id="ack"><title>Acknowledgements</title><p>This study was supported by grants to MEB and BAW from the NCI (5R01CA234904). ARL was supported by the UW Cellular and Molecular Pathology (5 T32GM081061) and the UW Genomic Sciences Training Program (5T32HG002760) NIH training grants. NLA was supported by T32GM008692. ASZ. was supported in part by T32GM008688.Technical support comes from University of Wisconsin Carbone Cancer Center (UWCCC) Shared Resources funded by the UWCCC Support Grant P30 CA014520 – Flow Cytometry Core Facility (1S10RR025483-01), Cancer Informatics Shared Resource, Small Molecule Screening Facility. The authors thank the UW Biotechnology Center DNA Sequencing Facility for providing Illumina sequencing services. Special thanks go to Drs. Ana Bolhaqueiro, Bas Ponsioen, and Geert Kops for the provision of scDNAseq data for our analyses and to Dr. Caitlin Pepperell for valuable comments related to approximate Bayesian computation.</p></ack><ref-list><title>References</title><ref id="bib1"><element-citation publication-type="preprint"><person-group person-group-type="author"><name><surname>Auslander</surname><given-names>N</given-names></name><name><surname>Heselmeyer-Haddad</surname><given-names>K</given-names></name><name><surname>Patkar</surname><given-names>S</given-names></name><name><surname>Hirsch</surname><given-names>D</given-names></name><name><surname>Camps</surname><given-names>J</given-names></name><name><surname>Brown</surname><given-names>M</given-names></name><name><surname>Bronder</surname><given-names>D</given-names></name><name><surname>Chen</surname><given-names>WD</given-names></name><name><surname>Lokanga</surname><given-names>R</given-names></name><name><surname>Wangsa</surname><given-names>D</given-names></name><name><surname>Wangsa</surname><given-names>D</given-names></name><name><surname>Hu</surname><given-names>Y</given-names></name><name><surname>Lischka</surname><given-names>A</given-names></name><name><surname>Braun</surname><given-names>R</given-names></name><name><surname>Emons</surname><given-names>G</given-names></name><name><surname>Ghadimi</surname><given-names>BM</given-names></name><name><surname>Gaedcke</surname><given-names>J</given-names></name><name><surname>Grade</surname><given-names>M</given-names></name><name><surname>Montagna</surname><given-names>C</given-names></name><name><surname>Lazebnik</surname><given-names>Y</given-names></name><name><surname>Difilippantonio</surname><given-names>MJ</given-names></name><name><surname>Habermann</surname><given-names>JK</given-names></name><name><surname>Auer</surname><given-names>G</given-names></name><name><surname>Ruppin</surname><given-names>E</given-names></name><name><surname>Ried</surname><given-names>T</given-names></name></person-group><year iso-8601-date="2019">2019</year><article-title>Cancer-Type Specific Aneuploidies Hard-Wire Chromosome-Wide Gene Expression Patterns of Their Tissue of Origin</article-title><source>bioRxiv</source><pub-id pub-id-type="doi">10.1101/563858</pub-id></element-citation></ref><ref id="bib2"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Bakhoum</surname><given-names>SF</given-names></name><name><surname>Thompson</surname><given-names>SL</given-names></name><name><surname>Manning</surname><given-names>AL</given-names></name><name><surname>Compton</surname><given-names>DA</given-names></name></person-group><year iso-8601-date="2009">2009</year><article-title>Genome stability is ensured by temporal control of kinetochore-microtubule dynamics</article-title><source>Nature Cell Biology</source><volume>11</volume><fpage>27</fpage><lpage>35</lpage><pub-id pub-id-type="doi">10.1038/ncb1809</pub-id><pub-id pub-id-type="pmid">19060894</pub-id></element-citation></ref><ref id="bib3"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Bakhoum</surname><given-names>SF</given-names></name><name><surname>Danilova</surname><given-names>OV</given-names></name><name><surname>Kaur</surname><given-names>P</given-names></name><name><surname>Levy</surname><given-names>NB</given-names></name><name><surname>Compton</surname><given-names>DA</given-names></name></person-group><year iso-8601-date="2011">2011</year><article-title>Chromosomal instability substantiates poor prognosis in patients with diffuse large B-cell lymphoma</article-title><source>Clinical Cancer Research</source><volume>17</volume><fpage>7704</fpage><lpage>7711</lpage><pub-id pub-id-type="doi">10.1158/1078-0432.CCR-11-2049</pub-id><pub-id pub-id-type="pmid">22184286</pub-id></element-citation></ref><ref id="bib4"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Bakhoum</surname><given-names>SF</given-names></name><name><surname>Silkworth</surname><given-names>WT</given-names></name><name><surname>Nardi</surname><given-names>IK</given-names></name><name><surname>Nicholson</surname><given-names>JM</given-names></name><name><surname>Compton</surname><given-names>DA</given-names></name><name><surname>Cimini</surname><given-names>D</given-names></name></person-group><year iso-8601-date="2014">2014</year><article-title>The mitotic origin of chromosomal instability</article-title><source>Current Biology: CB</source><volume>24</volume><fpage>R148</fpage><lpage>R149</lpage><pub-id pub-id-type="doi">10.1016/j.cub.2014.01.019</pub-id></element-citation></ref><ref id="bib5"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Bakhoum</surname><given-names>SF</given-names></name><name><surname>Ngo</surname><given-names>B</given-names></name><name><surname>Laughney</surname><given-names>AM</given-names></name><name><surname>Cavallo</surname><given-names>J-A</given-names></name><name><surname>Murphy</surname><given-names>CJ</given-names></name><name><surname>Ly</surname><given-names>P</given-names></name><name><surname>Shah</surname><given-names>P</given-names></name><name><surname>Sriram</surname><given-names>RK</given-names></name><name><surname>Watkins</surname><given-names>TBK</given-names></name><name><surname>Taunk</surname><given-names>NK</given-names></name><name><surname>Duran</surname><given-names>M</given-names></name><name><surname>Pauli</surname><given-names>C</given-names></name><name><surname>Shaw</surname><given-names>C</given-names></name><name><surname>Chadalavada</surname><given-names>K</given-names></name><name><surname>Rajasekhar</surname><given-names>VK</given-names></name><name><surname>Genovese</surname><given-names>G</given-names></name><name><surname>Venkatesan</surname><given-names>S</given-names></name><name><surname>Birkbak</surname><given-names>NJ</given-names></name><name><surname>McGranahan</surname><given-names>N</given-names></name><name><surname>Lundquist</surname><given-names>M</given-names></name><name><surname>LaPlant</surname><given-names>Q</given-names></name><name><surname>Healey</surname><given-names>JH</given-names></name><name><surname>Elemento</surname><given-names>O</given-names></name><name><surname>Chung</surname><given-names>CH</given-names></name><name><surname>Lee</surname><given-names>NY</given-names></name><name><surname>Imielenski</surname><given-names>M</given-names></name><name><surname>Nanjangud</surname><given-names>G</given-names></name><name><surname>Pe’er</surname><given-names>D</given-names></name><name><surname>Cleveland</surname><given-names>DW</given-names></name><name><surname>Powell</surname><given-names>SN</given-names></name><name><surname>Lammerding</surname><given-names>J</given-names></name><name><surname>Swanton</surname><given-names>C</given-names></name><name><surname>Cantley</surname><given-names>LC</given-names></name></person-group><year iso-8601-date="2018">2018</year><article-title>Chromosomal instability drives metastasis through a cytosolic DNA response</article-title><source>Nature</source><volume>553</volume><fpage>467</fpage><lpage>472</lpage><pub-id pub-id-type="doi">10.1038/nature25432</pub-id><pub-id pub-id-type="pmid">29342134</pub-id></element-citation></ref><ref id="bib6"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Baslan</surname><given-names>T</given-names></name><name><surname>Kendall</surname><given-names>J</given-names></name><name><surname>Ward</surname><given-names>B</given-names></name><name><surname>Cox</surname><given-names>H</given-names></name><name><surname>Leotta</surname><given-names>A</given-names></name><name><surname>Rodgers</surname><given-names>L</given-names></name><name><surname>Riggs</surname><given-names>M</given-names></name><name><surname>D’Italia</surname><given-names>S</given-names></name><name><surname>Sun</surname><given-names>G</given-names></name><name><surname>Yong</surname><given-names>M</given-names></name><name><surname>Miskimen</surname><given-names>K</given-names></name><name><surname>Gilmore</surname><given-names>H</given-names></name><name><surname>Saborowski</surname><given-names>M</given-names></name><name><surname>Dimitrova</surname><given-names>N</given-names></name><name><surname>Krasnitz</surname><given-names>A</given-names></name><name><surname>Harris</surname><given-names>L</given-names></name><name><surname>Wigler</surname><given-names>M</given-names></name><name><surname>Hicks</surname><given-names>J</given-names></name></person-group><year iso-8601-date="2015">2015</year><article-title>Optimizing sparse sequencing of single cells for highly multiplex copy number profiling</article-title><source>Genome Research</source><volume>25</volume><fpage>714</fpage><lpage>724</lpage><pub-id pub-id-type="doi">10.1101/gr.188060.114</pub-id><pub-id pub-id-type="pmid">25858951</pub-id></element-citation></ref><ref id="bib7"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Bielski</surname><given-names>CM</given-names></name><name><surname>Zehir</surname><given-names>A</given-names></name><name><surname>Penson</surname><given-names>AV</given-names></name><name><surname>Donoghue</surname><given-names>MTA</given-names></name><name><surname>Chatila</surname><given-names>W</given-names></name><name><surname>Armenia</surname><given-names>J</given-names></name><name><surname>Chang</surname><given-names>MT</given-names></name><name><surname>Schram</surname><given-names>AM</given-names></name><name><surname>Jonsson</surname><given-names>P</given-names></name><name><surname>Bandlamudi</surname><given-names>C</given-names></name><name><surname>Razavi</surname><given-names>P</given-names></name><name><surname>Iyer</surname><given-names>G</given-names></name><name><surname>Robson</surname><given-names>ME</given-names></name><name><surname>Stadler</surname><given-names>ZK</given-names></name><name><surname>Schultz</surname><given-names>N</given-names></name><name><surname>Baselga</surname><given-names>J</given-names></name><name><surname>Solit</surname><given-names>DB</given-names></name><name><surname>Hyman</surname><given-names>DM</given-names></name><name><surname>Berger</surname><given-names>MF</given-names></name><name><surname>Taylor</surname><given-names>BS</given-names></name></person-group><year iso-8601-date="2018">2018</year><article-title>Genome doubling shapes the evolution and prognosis of advanced cancers</article-title><source>Nature Genetics</source><volume>50</volume><fpage>1189</fpage><lpage>1195</lpage><pub-id pub-id-type="doi">10.1038/s41588-018-0165-1</pub-id><pub-id pub-id-type="pmid">30013179</pub-id></element-citation></ref><ref id="bib8"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Bolhaqueiro</surname><given-names>ACF</given-names></name><name><surname>Ponsioen</surname><given-names>B</given-names></name><name><surname>Bakker</surname><given-names>B</given-names></name><name><surname>Klaasen</surname><given-names>SJ</given-names></name><name><surname>Kucukkose</surname><given-names>E</given-names></name><name><surname>van Jaarsveld</surname><given-names>RH</given-names></name><name><surname>Vivié</surname><given-names>J</given-names></name><name><surname>Verlaan-Klink</surname><given-names>I</given-names></name><name><surname>Hami</surname><given-names>N</given-names></name><name><surname>Spierings</surname><given-names>DCJ</given-names></name><name><surname>Sasaki</surname><given-names>N</given-names></name><name><surname>Dutta</surname><given-names>D</given-names></name><name><surname>Boj</surname><given-names>SF</given-names></name><name><surname>Vries</surname><given-names>RGJ</given-names></name><name><surname>Lansdorp</surname><given-names>PM</given-names></name><name><surname>van de Wetering</surname><given-names>M</given-names></name><name><surname>van Oudenaarden</surname><given-names>A</given-names></name><name><surname>Clevers</surname><given-names>H</given-names></name><name><surname>Kranenburg</surname><given-names>O</given-names></name><name><surname>Foijer</surname><given-names>F</given-names></name><name><surname>Snippert</surname><given-names>HJG</given-names></name><name><surname>Kops</surname><given-names>GJPL</given-names></name></person-group><year iso-8601-date="2019">2019</year><article-title>Ongoing chromosomal instability and karyotype evolution in human colorectal cancer organoids</article-title><source>Nature Genetics</source><volume>51</volume><fpage>824</fpage><lpage>834</lpage><pub-id pub-id-type="doi">10.1038/s41588-019-0399-6</pub-id><pub-id pub-id-type="pmid">31036964</pub-id></element-citation></ref><ref id="bib9"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Bollen</surname><given-names>Y</given-names></name><name><surname>Stelloo</surname><given-names>E</given-names></name><name><surname>van Leenen</surname><given-names>P</given-names></name><name><surname>van den Bos</surname><given-names>M</given-names></name><name><surname>Ponsioen</surname><given-names>B</given-names></name><name><surname>Lu</surname><given-names>B</given-names></name><name><surname>van Roosmalen</surname><given-names>MJ</given-names></name><name><surname>Bolhaqueiro</surname><given-names>ACF</given-names></name><name><surname>Kimberley</surname><given-names>C</given-names></name><name><surname>Mossner</surname><given-names>M</given-names></name><name><surname>Cross</surname><given-names>WCH</given-names></name><name><surname>Besselink</surname><given-names>NJM</given-names></name><name><surname>van der Roest</surname><given-names>B</given-names></name><name><surname>Boymans</surname><given-names>S</given-names></name><name><surname>Oost</surname><given-names>KC</given-names></name><name><surname>de Vries</surname><given-names>SG</given-names></name><name><surname>Rehmann</surname><given-names>H</given-names></name><name><surname>Cuppen</surname><given-names>E</given-names></name><name><surname>Lens</surname><given-names>SMA</given-names></name><name><surname>Kops</surname><given-names>GJPL</given-names></name><name><surname>Kloosterman</surname><given-names>WP</given-names></name><name><surname>Terstappen</surname><given-names>LWMM</given-names></name><name><surname>Barnes</surname><given-names>CP</given-names></name><name><surname>Sottoriva</surname><given-names>A</given-names></name><name><surname>Graham</surname><given-names>TA</given-names></name><name><surname>Snippert</surname><given-names>HJG</given-names></name></person-group><year iso-8601-date="2021">2021</year><article-title>Reconstructing single-cell karyotype alterations in colorectal cancer identifies punctuated and gradual diversification patterns</article-title><source>Nature Genetics</source><volume>53</volume><fpage>1187</fpage><lpage>1195</lpage><pub-id pub-id-type="doi">10.1038/s41588-021-00891-2</pub-id><pub-id pub-id-type="pmid">34211178</pub-id></element-citation></ref><ref id="bib10"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Carter</surname><given-names>SL</given-names></name><name><surname>Eklund</surname><given-names>AC</given-names></name><name><surname>Kohane</surname><given-names>IS</given-names></name><name><surname>Harris</surname><given-names>LN</given-names></name><name><surname>Szallasi</surname><given-names>Z</given-names></name></person-group><year iso-8601-date="2006">2006</year><article-title>A signature of chromosomal instability inferred from gene expression profiles predicts clinical outcome in multiple human cancers</article-title><source>Nature Genetics</source><volume>38</volume><fpage>1043</fpage><lpage>1048</lpage><pub-id pub-id-type="doi">10.1038/ng1861</pub-id><pub-id pub-id-type="pmid">16921376</pub-id></element-citation></ref><ref id="bib11"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Carter</surname><given-names>SL</given-names></name><name><surname>Cibulskis</surname><given-names>K</given-names></name><name><surname>Helman</surname><given-names>E</given-names></name><name><surname>McKenna</surname><given-names>A</given-names></name><name><surname>Shen</surname><given-names>H</given-names></name><name><surname>Zack</surname><given-names>T</given-names></name><name><surname>Laird</surname><given-names>PW</given-names></name><name><surname>Onofrio</surname><given-names>RC</given-names></name><name><surname>Winckler</surname><given-names>W</given-names></name><name><surname>Weir</surname><given-names>BA</given-names></name><name><surname>Beroukhim</surname><given-names>R</given-names></name><name><surname>Pellman</surname><given-names>D</given-names></name><name><surname>Levine</surname><given-names>DA</given-names></name><name><surname>Lander</surname><given-names>ES</given-names></name><name><surname>Meyerson</surname><given-names>M</given-names></name><name><surname>Getz</surname><given-names>G</given-names></name></person-group><year iso-8601-date="2012">2012</year><article-title>Absolute quantification of somatic DNA alterations in human cancer</article-title><source>Nature Biotechnology</source><volume>30</volume><fpage>413</fpage><lpage>421</lpage><pub-id pub-id-type="doi">10.1038/nbt.2203</pub-id><pub-id pub-id-type="pmid">22544022</pub-id></element-citation></ref><ref id="bib12"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Colijn</surname><given-names>C</given-names></name><name><surname>Plazzotta</surname><given-names>G</given-names></name></person-group><year iso-8601-date="2018">2018</year><article-title>A Metric on Phylogenetic Tree Shapes</article-title><source>Systematic Biology</source><volume>67</volume><fpage>113</fpage><lpage>126</lpage><pub-id pub-id-type="doi">10.1093/sysbio/syx046</pub-id><pub-id pub-id-type="pmid">28472435</pub-id></element-citation></ref><ref id="bib13"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Csilléry</surname><given-names>K</given-names></name><name><surname>Blum</surname><given-names>MGB</given-names></name><name><surname>Gaggiotti</surname><given-names>OE</given-names></name><name><surname>François</surname><given-names>O</given-names></name></person-group><year iso-8601-date="2010">2010</year><article-title>Approximate Bayesian Computation (ABC) in practice</article-title><source>Trends in Ecology & Evolution</source><volume>25</volume><fpage>410</fpage><lpage>418</lpage><pub-id pub-id-type="doi">10.1016/j.tree.2010.04.001</pub-id><pub-id pub-id-type="pmid">20488578</pub-id></element-citation></ref><ref id="bib14"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Csilléry</surname><given-names>K</given-names></name><name><surname>François</surname><given-names>O</given-names></name><name><surname>Blum</surname><given-names>MGB</given-names></name></person-group><year iso-8601-date="2012">2012</year><article-title>ABC: an R package for approximate Bayesian computation (ABC</article-title><source>Methods in Ecology and Evolution</source><volume>3</volume><fpage>475</fpage><lpage>479</lpage><pub-id pub-id-type="doi">10.1111/j.2041-210X.2011.00179.x</pub-id></element-citation></ref><ref id="bib15"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Davoli</surname><given-names>T</given-names></name><name><surname>Xu</surname><given-names>AW</given-names></name><name><surname>Mengwasser</surname><given-names>KE</given-names></name><name><surname>Sack</surname><given-names>LM</given-names></name><name><surname>Yoon</surname><given-names>JC</given-names></name><name><surname>Park</surname><given-names>PJ</given-names></name><name><surname>Elledge</surname><given-names>SJ</given-names></name></person-group><year iso-8601-date="2013">2013</year><article-title>Cumulative haploinsufficiency and triplosensitivity drive aneuploidy patterns and shape the cancer genome</article-title><source>Cell</source><volume>155</volume><fpage>948</fpage><lpage>962</lpage><pub-id pub-id-type="doi">10.1016/j.cell.2013.10.011</pub-id><pub-id pub-id-type="pmid">24183448</pub-id></element-citation></ref><ref id="bib16"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Davoli</surname><given-names>T</given-names></name><name><surname>Uno</surname><given-names>H</given-names></name><name><surname>Wooten</surname><given-names>EC</given-names></name><name><surname>Elledge</surname><given-names>SJ</given-names></name></person-group><year iso-8601-date="2017">2017</year><article-title>Tumor aneuploidy correlates with markers of immune evasion and with reduced response to immunotherapy</article-title><source>Science (New York, N.Y.)</source><volume>355</volume><elocation-id>eaaf8399</elocation-id><pub-id pub-id-type="doi">10.1126/science.aaf8399</pub-id><pub-id pub-id-type="pmid">28104840</pub-id></element-citation></ref><ref id="bib17"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Dayarian</surname><given-names>A</given-names></name><name><surname>Shraiman</surname><given-names>BI</given-names></name></person-group><year iso-8601-date="2014">2014</year><article-title>How to infer relative fitness from a sample of genomic sequences</article-title><source>Genetics</source><volume>197</volume><fpage>913</fpage><lpage>923</lpage><pub-id pub-id-type="doi">10.1534/genetics.113.160986</pub-id><pub-id pub-id-type="pmid">24770330</pub-id></element-citation></ref><ref id="bib18"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Denu</surname><given-names>RA</given-names></name><name><surname>Zasadil</surname><given-names>LM</given-names></name><name><surname>Kanugh</surname><given-names>C</given-names></name><name><surname>Laffin</surname><given-names>J</given-names></name><name><surname>Weaver</surname><given-names>BA</given-names></name><name><surname>Burkard</surname><given-names>ME</given-names></name></person-group><year iso-8601-date="2016">2016</year><article-title>Centrosome amplification induces high grade features and is prognostic of worse outcomes in breast cancer</article-title><source>BMC Cancer</source><volume>16</volume><fpage>1</fpage><lpage>13</lpage><pub-id pub-id-type="doi">10.1186/s12885-016-2083-x</pub-id><pub-id pub-id-type="pmid">26832928</pub-id></element-citation></ref><ref id="bib19"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Dewhurst</surname><given-names>SM</given-names></name><name><surname>McGranahan</surname><given-names>N</given-names></name><name><surname>Burrell</surname><given-names>RA</given-names></name><name><surname>Rowan</surname><given-names>AJ</given-names></name><name><surname>Grönroos</surname><given-names>E</given-names></name><name><surname>Endesfelder</surname><given-names>D</given-names></name><name><surname>Joshi</surname><given-names>T</given-names></name><name><surname>Mouradov</surname><given-names>D</given-names></name><name><surname>Gibbs</surname><given-names>P</given-names></name><name><surname>Ward</surname><given-names>RL</given-names></name><name><surname>Hawkins</surname><given-names>NJ</given-names></name><name><surname>Szallasi</surname><given-names>Z</given-names></name><name><surname>Sieber</surname><given-names>OM</given-names></name><name><surname>Swanton</surname><given-names>C</given-names></name></person-group><year iso-8601-date="2014">2014</year><article-title>Tolerance of whole-genome doubling propagates chromosomal instability and accelerates cancer genome evolution</article-title><source>Cancer Discovery</source><volume>4</volume><fpage>175</fpage><lpage>185</lpage><pub-id pub-id-type="doi">10.1158/2159-8290.CD-13-0285</pub-id><pub-id pub-id-type="pmid">24436049</pub-id></element-citation></ref><ref id="bib20"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Dumont</surname><given-names>M</given-names></name><name><surname>Gamba</surname><given-names>R</given-names></name><name><surname>Gestraud</surname><given-names>P</given-names></name><name><surname>Klaasen</surname><given-names>S</given-names></name><name><surname>Worrall</surname><given-names>JT</given-names></name><name><surname>De Vries</surname><given-names>SG</given-names></name><name><surname>Boudreau</surname><given-names>V</given-names></name><name><surname>Salinas-Luypaert</surname><given-names>C</given-names></name><name><surname>Maddox</surname><given-names>PS</given-names></name><name><surname>Lens</surname><given-names>SM</given-names></name><name><surname>Kops</surname><given-names>GJ</given-names></name><name><surname>McClelland</surname><given-names>SE</given-names></name><name><surname>Miga</surname><given-names>KH</given-names></name><name><surname>Fachinetti</surname><given-names>D</given-names></name></person-group><year iso-8601-date="2020">2020</year><article-title>Human chromosome-specific aneuploidy is influenced by DNA-dependent centromeric features</article-title><source>The EMBO Journal</source><volume>39</volume><elocation-id>e102924</elocation-id><pub-id pub-id-type="doi">10.15252/embj.2019102924</pub-id><pub-id pub-id-type="pmid">31750958</pub-id></element-citation></ref><ref id="bib21"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Dürrbaum</surname><given-names>M</given-names></name><name><surname>Kuznetsova</surname><given-names>AY</given-names></name><name><surname>Passerini</surname><given-names>V</given-names></name><name><surname>Stingele</surname><given-names>S</given-names></name><name><surname>Stoehr</surname><given-names>G</given-names></name><name><surname>Storchová</surname><given-names>Z</given-names></name></person-group><year iso-8601-date="2014">2014</year><article-title>Unique features of the transcriptional response to model aneuploidy in human cells</article-title><source>BMC Genomics</source><volume>15</volume><elocation-id>139</elocation-id><pub-id pub-id-type="doi">10.1186/1471-2164-15-139</pub-id><pub-id pub-id-type="pmid">24548329</pub-id></element-citation></ref><ref id="bib22"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Elizalde</surname><given-names>S</given-names></name><name><surname>Laughney</surname><given-names>AM</given-names></name><name><surname>Bakhoum</surname><given-names>SF</given-names></name></person-group><year iso-8601-date="2018">2018</year><article-title>A Markov chain for numerical chromosomal instability in clonally expanding populations</article-title><source>PLOS Computational Biology</source><volume>14</volume><elocation-id>e1006447</elocation-id><pub-id pub-id-type="doi">10.1371/journal.pcbi.1006447</pub-id><pub-id pub-id-type="pmid">30204765</pub-id></element-citation></ref><ref id="bib23"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Evrony</surname><given-names>GD</given-names></name><name><surname>Hinch</surname><given-names>AG</given-names></name><name><surname>Luo</surname><given-names>C</given-names></name></person-group><year iso-8601-date="2021">2021</year><article-title>Applications of Single-Cell DNA Sequencing</article-title><source>Annual Review of Genomics and Human Genetics</source><volume>22</volume><fpage>171</fpage><lpage>197</lpage><pub-id pub-id-type="doi">10.1146/annurev-genom-111320-090436</pub-id><pub-id pub-id-type="pmid">33722077</pub-id></element-citation></ref><ref id="bib24"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Foijer</surname><given-names>F</given-names></name><name><surname>Xie</surname><given-names>SZ</given-names></name><name><surname>Simon</surname><given-names>JE</given-names></name><name><surname>Bakker</surname><given-names>PL</given-names></name><name><surname>Conte</surname><given-names>N</given-names></name><name><surname>Davis</surname><given-names>SH</given-names></name><name><surname>Kregel</surname><given-names>E</given-names></name><name><surname>Jonkers</surname><given-names>J</given-names></name><name><surname>Bradley</surname><given-names>A</given-names></name><name><surname>Sorger</surname><given-names>PK</given-names></name></person-group><year iso-8601-date="2014">2014</year><article-title>Chromosome instability induced by Mps1 and p53 mutation generates aggressive lymphomas exhibiting aneuploidy-induced stress</article-title><source>PNAS</source><volume>111</volume><fpage>13427</fpage><lpage>13432</lpage><pub-id pub-id-type="doi">10.1073/pnas.1400892111</pub-id><pub-id pub-id-type="pmid">25197064</pub-id></element-citation></ref><ref id="bib25"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Foijer</surname><given-names>F</given-names></name><name><surname>Albacker</surname><given-names>LA</given-names></name><name><surname>Bakker</surname><given-names>B</given-names></name><name><surname>Spierings</surname><given-names>DC</given-names></name><name><surname>Yue</surname><given-names>Y</given-names></name><name><surname>Xie</surname><given-names>SZ</given-names></name><name><surname>Davis</surname><given-names>S</given-names></name><name><surname>Lutum-Jehle</surname><given-names>A</given-names></name><name><surname>Takemoto</surname><given-names>D</given-names></name><name><surname>Hare</surname><given-names>B</given-names></name><name><surname>Furey</surname><given-names>B</given-names></name><name><surname>Bronson</surname><given-names>RT</given-names></name><name><surname>Lansdorp</surname><given-names>PM</given-names></name><name><surname>Bradley</surname><given-names>A</given-names></name><name><surname>Sorger</surname><given-names>PK</given-names></name></person-group><year iso-8601-date="2017">2017</year><article-title>Deletion of the <italic>MAD2L1</italic> spindle assembly checkpoint gene is tolerated in mouse models of acute T-cell lymphoma and hepatocellular carcinoma</article-title><source>eLife</source><volume>6</volume><elocation-id>e20873</elocation-id><pub-id pub-id-type="doi">10.7554/eLife.20873</pub-id><pub-id pub-id-type="pmid">28318489</pub-id></element-citation></ref><ref id="bib26"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Funk</surname><given-names>LC</given-names></name><name><surname>Wan</surname><given-names>J</given-names></name><name><surname>Ryan</surname><given-names>SD</given-names></name><name><surname>Kaur</surname><given-names>C</given-names></name><name><surname>Sullivan</surname><given-names>R</given-names></name><name><surname>Roopra</surname><given-names>A</given-names></name><name><surname>Weaver</surname><given-names>BA</given-names></name></person-group><year iso-8601-date="2021">2021</year><article-title>p53 Is Not Required for High CIN to Induce Tumor Suppression</article-title><source>Molecular Cancer Research</source><volume>19</volume><fpage>112</fpage><lpage>123</lpage><pub-id pub-id-type="doi">10.1158/1541-7786.MCR-20-0488</pub-id><pub-id pub-id-type="pmid">32948674</pub-id></element-citation></ref><ref id="bib27"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Gao</surname><given-names>R</given-names></name><name><surname>Davis</surname><given-names>A</given-names></name><name><surname>McDonald</surname><given-names>TO</given-names></name><name><surname>Sei</surname><given-names>E</given-names></name><name><surname>Shi</surname><given-names>X</given-names></name><name><surname>Wang</surname><given-names>Y</given-names></name><name><surname>Tsai</surname><given-names>PC</given-names></name><name><surname>Casasent</surname><given-names>A</given-names></name><name><surname>Waters</surname><given-names>J</given-names></name><name><surname>Zhang</surname><given-names>H</given-names></name><name><surname>Meric-Bernstam</surname><given-names>F</given-names></name><name><surname>Michor</surname><given-names>F</given-names></name><name><surname>Navin</surname><given-names>NE</given-names></name></person-group><year iso-8601-date="2016">2016</year><article-title>Punctuated copy number evolution and clonal stasis in triple-negative breast cancer</article-title><source>Nature Genetics</source><volume>48</volume><fpage>1119</fpage><lpage>1130</lpage><pub-id pub-id-type="doi">10.1038/ng.3641</pub-id><pub-id pub-id-type="pmid">27526321</pub-id></element-citation></ref><ref id="bib28"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Garvin</surname><given-names>T</given-names></name><name><surname>Aboukhalil</surname><given-names>R</given-names></name><name><surname>Kendall</surname><given-names>J</given-names></name><name><surname>Baslan</surname><given-names>T</given-names></name><name><surname>Atwal</surname><given-names>GS</given-names></name><name><surname>Hicks</surname><given-names>J</given-names></name><name><surname>Wigler</surname><given-names>M</given-names></name><name><surname>Schatz</surname><given-names>MC</given-names></name></person-group><year iso-8601-date="2015">2015</year><article-title>Interactive analysis and assessment of single-cell copy-number variations</article-title><source>Nature Methods</source><volume>12</volume><fpage>1058</fpage><lpage>1060</lpage><pub-id pub-id-type="doi">10.1038/nmeth.3578</pub-id><pub-id pub-id-type="pmid">26344043</pub-id></element-citation></ref><ref id="bib29"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Gerstung</surname><given-names>M</given-names></name><name><surname>Jolly</surname><given-names>C</given-names></name><name><surname>Leshchiner</surname><given-names>I</given-names></name><name><surname>Dentro</surname><given-names>SC</given-names></name><name><surname>Gonzalez</surname><given-names>S</given-names></name><name><surname>Rosebrock</surname><given-names>D</given-names></name><name><surname>Mitchell</surname><given-names>TJ</given-names></name><name><surname>Rubanova</surname><given-names>Y</given-names></name><name><surname>Anur</surname><given-names>P</given-names></name><name><surname>Yu</surname><given-names>K</given-names></name><name><surname>Tarabichi</surname><given-names>M</given-names></name><name><surname>Deshwar</surname><given-names>A</given-names></name><name><surname>Wintersinger</surname><given-names>J</given-names></name><name><surname>Kleinheinz</surname><given-names>K</given-names></name><name><surname>Vázquez-García</surname><given-names>I</given-names></name><name><surname>Haase</surname><given-names>K</given-names></name><name><surname>Jerman</surname><given-names>L</given-names></name><name><surname>Sengupta</surname><given-names>S</given-names></name><name><surname>Macintyre</surname><given-names>G</given-names></name><name><surname>Malikic</surname><given-names>S</given-names></name><name><surname>Donmez</surname><given-names>N</given-names></name><name><surname>Livitz</surname><given-names>DG</given-names></name><name><surname>Cmero</surname><given-names>M</given-names></name><name><surname>Demeulemeester</surname><given-names>J</given-names></name><name><surname>Schumacher</surname><given-names>S</given-names></name><name><surname>Fan</surname><given-names>Y</given-names></name><name><surname>Yao</surname><given-names>X</given-names></name><name><surname>Lee</surname><given-names>J</given-names></name><name><surname>Schlesner</surname><given-names>M</given-names></name><name><surname>Boutros</surname><given-names>PC</given-names></name><name><surname>Bowtell</surname><given-names>DD</given-names></name><name><surname>Zhu</surname><given-names>H</given-names></name><name><surname>Getz</surname><given-names>G</given-names></name><name><surname>Imielinski</surname><given-names>M</given-names></name><name><surname>Beroukhim</surname><given-names>R</given-names></name><name><surname>Sahinalp</surname><given-names>SC</given-names></name><name><surname>Ji</surname><given-names>Y</given-names></name><name><surname>Peifer</surname><given-names>M</given-names></name><name><surname>Markowetz</surname><given-names>F</given-names></name><name><surname>Mustonen</surname><given-names>V</given-names></name><name><surname>Yuan</surname><given-names>K</given-names></name><name><surname>Wang</surname><given-names>W</given-names></name><name><surname>Morris</surname><given-names>QD</given-names></name><name><surname>Spellman</surname><given-names>PT</given-names></name><name><surname>Wedge</surname><given-names>DC</given-names></name><name><surname>Van Loo</surname><given-names>P</given-names></name><collab>PCAWG Evolution & Heterogeneity Working Group</collab><collab>PCAWG Consortium</collab></person-group><year iso-8601-date="2020">2020</year><article-title>The evolutionary history of 2,658 cancers</article-title><source>Nature</source><volume>578</volume><fpage>122</fpage><lpage>128</lpage><pub-id pub-id-type="doi">10.1038/s41586-019-1907-7</pub-id><pub-id pub-id-type="pmid">32025013</pub-id></element-citation></ref><ref id="bib30"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Grim</surname><given-names>JE</given-names></name><name><surname>Knoblaugh</surname><given-names>SE</given-names></name><name><surname>Guthrie</surname><given-names>KA</given-names></name><name><surname>Hagar</surname><given-names>A</given-names></name><name><surname>Swanger</surname><given-names>J</given-names></name><name><surname>Hespelt</surname><given-names>J</given-names></name><name><surname>Delrow</surname><given-names>JJ</given-names></name><name><surname>Small</surname><given-names>T</given-names></name><name><surname>Grady</surname><given-names>WM</given-names></name><name><surname>Nakayama</surname><given-names>KI</given-names></name><name><surname>Clurman</surname><given-names>BE</given-names></name></person-group><year iso-8601-date="2012">2012</year><article-title>Fbw7 and p53 cooperatively suppress advanced and chromosomally unstable intestinal cancer</article-title><source>Molecular and Cellular Biology</source><volume>32</volume><fpage>2160</fpage><lpage>2167</lpage><pub-id pub-id-type="doi">10.1128/MCB.00305-12</pub-id><pub-id pub-id-type="pmid">22473991</pub-id></element-citation></ref><ref id="bib31"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Gusev</surname><given-names>Y</given-names></name><name><surname>Kagansky</surname><given-names>V</given-names></name><name><surname>Dooley</surname><given-names>WC</given-names></name></person-group><year iso-8601-date="2000">2000</year><article-title>A stochastic model of chromosome segregation errors with reference to cancer cells</article-title><source>Mathematical and Computer Modelling</source><volume>32</volume><fpage>97</fpage><lpage>111</lpage><pub-id pub-id-type="doi">10.1016/S0895-7177(00)00122-9</pub-id></element-citation></ref><ref id="bib32"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Gusev</surname><given-names>Y</given-names></name><name><surname>Kagansky</surname><given-names>V</given-names></name><name><surname>Dooley</surname><given-names>WC</given-names></name></person-group><year iso-8601-date="2001">2001</year><article-title>Long-term dynamics of chromosomal instability in cancer: A transition probability model</article-title><source>Mathematical and Computer Modelling</source><volume>33</volume><fpage>1253</fpage><lpage>1273</lpage><pub-id pub-id-type="doi">10.1016/S0895-7177(00)00313-7</pub-id></element-citation></ref><ref id="bib33"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Hanahan</surname><given-names>D</given-names></name><name><surname>Weinberg</surname><given-names>RA</given-names></name></person-group><year iso-8601-date="2011">2011</year><article-title>Hallmarks of cancer: the next generation</article-title><source>Cell</source><volume>144</volume><fpage>646</fpage><lpage>674</lpage><pub-id pub-id-type="doi">10.1016/j.cell.2011.02.013</pub-id><pub-id pub-id-type="pmid">21376230</pub-id></element-citation></ref><ref id="bib34"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Hancock</surname><given-names>JM</given-names></name><name><surname>Zvelebil</surname><given-names>MJ</given-names></name><name><surname>Griffith</surname><given-names>M</given-names></name><name><surname>Griffith</surname><given-names>OL</given-names></name></person-group><year iso-8601-date="2004">2004</year><article-title>Mitelman database of chromosome aberrations and gene fusions in cancer</article-title><source>Dictionary of Bioinformatics and Computational Biology</source><volume>2004</volume><elocation-id>0996</elocation-id><pub-id pub-id-type="doi">10.1002/9780471650126.dob0996</pub-id></element-citation></ref><ref id="bib35"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Hoevenaar</surname><given-names>WHM</given-names></name><name><surname>Janssen</surname><given-names>A</given-names></name><name><surname>Quirindongo</surname><given-names>AI</given-names></name><name><surname>Ma</surname><given-names>H</given-names></name><name><surname>Klaasen</surname><given-names>SJ</given-names></name><name><surname>Teixeira</surname><given-names>A</given-names></name><name><surname>van Gerwen</surname><given-names>B</given-names></name><name><surname>Lansu</surname><given-names>N</given-names></name><name><surname>Morsink</surname><given-names>FHM</given-names></name><name><surname>Offerhaus</surname><given-names>GJA</given-names></name><name><surname>Medema</surname><given-names>RH</given-names></name><name><surname>Kops</surname><given-names>GJPL</given-names></name><name><surname>Jelluma</surname><given-names>N</given-names></name></person-group><year iso-8601-date="2020">2020</year><article-title>Degree and site of chromosomal instability define its oncogenic potential</article-title><source>Nature Communications</source><volume>11</volume><elocation-id>1501</elocation-id><pub-id pub-id-type="doi">10.1038/s41467-020-15279-9</pub-id><pub-id pub-id-type="pmid">32198375</pub-id></element-citation></ref><ref id="bib36"><element-citation publication-type="preprint"><person-group person-group-type="author"><name><surname>Ippolito</surname><given-names>MR</given-names></name><name><surname>Martis</surname><given-names>V</given-names></name><name><surname>Hong</surname><given-names>C</given-names></name><name><surname>Wardenaar</surname><given-names>R</given-names></name><name><surname>Zerbib</surname><given-names>J</given-names></name><name><surname>Spierings</surname><given-names>DCJ</given-names></name><name><surname>Ben-David</surname><given-names>U</given-names></name><name><surname>Foijer</surname><given-names>F</given-names></name><name><surname>Santaguida</surname><given-names>S</given-names></name></person-group><year iso-8601-date="2020">2020</year><article-title>Aneuploidy-Driven Genome Instability Triggers Resistance to Chemotherapy</article-title><source>bioRxiv</source><pub-id pub-id-type="doi">10.1101/2020.09.25.313924</pub-id></element-citation></ref><ref id="bib37"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Jamal-Hanjani</surname><given-names>M</given-names></name><name><surname>Wilson</surname><given-names>GA</given-names></name><name><surname>McGranahan</surname><given-names>N</given-names></name><name><surname>Birkbak</surname><given-names>NJ</given-names></name><name><surname>Watkins</surname><given-names>TBK</given-names></name><name><surname>Veeriah</surname><given-names>S</given-names></name><name><surname>Shafi</surname><given-names>S</given-names></name><name><surname>Johnson</surname><given-names>DH</given-names></name><name><surname>Mitter</surname><given-names>R</given-names></name><name><surname>Rosenthal</surname><given-names>R</given-names></name><name><surname>Salm</surname><given-names>M</given-names></name><name><surname>Horswell</surname><given-names>S</given-names></name><name><surname>Escudero</surname><given-names>M</given-names></name><name><surname>Matthews</surname><given-names>N</given-names></name><name><surname>Rowan</surname><given-names>A</given-names></name><name><surname>Chambers</surname><given-names>T</given-names></name><name><surname>Moore</surname><given-names>DA</given-names></name><name><surname>Turajlic</surname><given-names>S</given-names></name><name><surname>Xu</surname><given-names>H</given-names></name><name><surname>Lee</surname><given-names>S-M</given-names></name><name><surname>Forster</surname><given-names>MD</given-names></name><name><surname>Ahmad</surname><given-names>T</given-names></name><name><surname>Hiley</surname><given-names>CT</given-names></name><name><surname>Abbosh</surname><given-names>C</given-names></name><name><surname>Falzon</surname><given-names>M</given-names></name><name><surname>Borg</surname><given-names>E</given-names></name><name><surname>Marafioti</surname><given-names>T</given-names></name><name><surname>Lawrence</surname><given-names>D</given-names></name><name><surname>Hayward</surname><given-names>M</given-names></name><name><surname>Kolvekar</surname><given-names>S</given-names></name><name><surname>Panagiotopoulos</surname><given-names>N</given-names></name><name><surname>Janes</surname><given-names>SM</given-names></name><name><surname>Thakrar</surname><given-names>R</given-names></name><name><surname>Ahmed</surname><given-names>A</given-names></name><name><surname>Blackhall</surname><given-names>F</given-names></name><name><surname>Summers</surname><given-names>Y</given-names></name><name><surname>Shah</surname><given-names>R</given-names></name><name><surname>Joseph</surname><given-names>L</given-names></name><name><surname>Quinn</surname><given-names>AM</given-names></name><name><surname>Crosbie</surname><given-names>PA</given-names></name><name><surname>Naidu</surname><given-names>B</given-names></name><name><surname>Middleton</surname><given-names>G</given-names></name><name><surname>Langman</surname><given-names>G</given-names></name><name><surname>Trotter</surname><given-names>S</given-names></name><name><surname>Nicolson</surname><given-names>M</given-names></name><name><surname>Remmen</surname><given-names>H</given-names></name><name><surname>Kerr</surname><given-names>K</given-names></name><name><surname>Chetty</surname><given-names>M</given-names></name><name><surname>Gomersall</surname><given-names>L</given-names></name><name><surname>Fennell</surname><given-names>DA</given-names></name><name><surname>Nakas</surname><given-names>A</given-names></name><name><surname>Rathinam</surname><given-names>S</given-names></name><name><surname>Anand</surname><given-names>G</given-names></name><name><surname>Khan</surname><given-names>S</given-names></name><name><surname>Russell</surname><given-names>P</given-names></name><name><surname>Ezhil</surname><given-names>V</given-names></name><name><surname>Ismail</surname><given-names>B</given-names></name><name><surname>Irvin-Sellers</surname><given-names>M</given-names></name><name><surname>Prakash</surname><given-names>V</given-names></name><name><surname>Lester</surname><given-names>JF</given-names></name><name><surname>Kornaszewska</surname><given-names>M</given-names></name><name><surname>Attanoos</surname><given-names>R</given-names></name><name><surname>Adams</surname><given-names>H</given-names></name><name><surname>Davies</surname><given-names>H</given-names></name><name><surname>Dentro</surname><given-names>S</given-names></name><name><surname>Taniere</surname><given-names>P</given-names></name><name><surname>O’Sullivan</surname><given-names>B</given-names></name><name><surname>Lowe</surname><given-names>HL</given-names></name><name><surname>Hartley</surname><given-names>JA</given-names></name><name><surname>Iles</surname><given-names>N</given-names></name><name><surname>Bell</surname><given-names>H</given-names></name><name><surname>Ngai</surname><given-names>Y</given-names></name><name><surname>Shaw</surname><given-names>JA</given-names></name><name><surname>Herrero</surname><given-names>J</given-names></name><name><surname>Szallasi</surname><given-names>Z</given-names></name><name><surname>Schwarz</surname><given-names>RF</given-names></name><name><surname>Stewart</surname><given-names>A</given-names></name><name><surname>Quezada</surname><given-names>SA</given-names></name><name><surname>Le Quesne</surname><given-names>J</given-names></name><name><surname>Van Loo</surname><given-names>P</given-names></name><name><surname>Dive</surname><given-names>C</given-names></name><name><surname>Hackshaw</surname><given-names>A</given-names></name><name><surname>Swanton</surname><given-names>C</given-names></name><collab>TRACERx Consortium</collab></person-group><year iso-8601-date="2017">2017</year><article-title>Tracking the Evolution of Non-Small-Cell Lung Cancer</article-title><source>The New England Journal of Medicine</source><volume>376</volume><fpage>2109</fpage><lpage>2121</lpage><pub-id pub-id-type="doi">10.1056/NEJMoa1616288</pub-id><pub-id pub-id-type="pmid">28445112</pub-id></element-citation></ref><ref id="bib38"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Janssen</surname><given-names>A</given-names></name><name><surname>Kops</surname><given-names>G</given-names></name><name><surname>Medema</surname><given-names>RH</given-names></name></person-group><year iso-8601-date="2009">2009</year><article-title>Elevating the frequency of chromosome mis-segregation as a strategy to kill tumor cells</article-title><source>PNAS</source><volume>106</volume><fpage>19108</fpage><lpage>19113</lpage><pub-id pub-id-type="doi">10.1073/pnas.0904343106</pub-id></element-citation></ref><ref id="bib39"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Jin</surname><given-names>N</given-names></name><name><surname>Lera</surname><given-names>RF</given-names></name><name><surname>Yan</surname><given-names>RE</given-names></name><name><surname>Guo</surname><given-names>F</given-names></name><name><surname>Oxendine</surname><given-names>K</given-names></name><name><surname>Horner</surname><given-names>VL</given-names></name><name><surname>Hu</surname><given-names>Y</given-names></name><name><surname>Wan</surname><given-names>J</given-names></name><name><surname>Mattison</surname><given-names>RJ</given-names></name><name><surname>Weaver</surname><given-names>BA</given-names></name><name><surname>Burkard</surname><given-names>ME</given-names></name></person-group><year iso-8601-date="2020">2020</year><article-title>Chromosomal instability upregulates interferon in acute myeloid leukemia</article-title><source>Genes, Chromosomes & Cancer</source><volume>59</volume><fpage>627</fpage><lpage>638</lpage><pub-id pub-id-type="doi">10.1002/gcc.22880</pub-id><pub-id pub-id-type="pmid">32557940</pub-id></element-citation></ref><ref id="bib40"><element-citation publication-type="software"><person-group person-group-type="author"><name><surname>Kendall</surname><given-names>M</given-names></name><name><surname>Boyd</surname><given-names>M</given-names></name><name><surname>Colijn</surname><given-names>C</given-names></name></person-group><year iso-8601-date="2018">2018</year><data-title>phyloTop: Tools for calculating and viewing topological properties of phylogenetic trees</data-title><version designator="1.4.1">1.4.1</version><publisher-name>Cran</publisher-name><ext-link ext-link-type="uri" xlink:href="https://cran.r-project.org/web/packages/phyloTop/README.html">https://cran.r-project.org/web/packages/phyloTop/README.html</ext-link></element-citation></ref><ref id="bib41"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Kim</surname><given-names>C</given-names></name><name><surname>Gao</surname><given-names>R</given-names></name><name><surname>Sei</surname><given-names>E</given-names></name><name><surname>Brandt</surname><given-names>R</given-names></name><name><surname>Hartman</surname><given-names>J</given-names></name><name><surname>Hatschek</surname><given-names>T</given-names></name><name><surname>Crosetto</surname><given-names>N</given-names></name><name><surname>Foukakis</surname><given-names>T</given-names></name><name><surname>Navin</surname><given-names>NE</given-names></name></person-group><year iso-8601-date="2018">2018</year><article-title>Chemoresistance Evolution in Triple-Negative Breast Cancer Delineated by Single-Cell Sequencing</article-title><source>Cell</source><volume>173</volume><fpage>879</fpage><lpage>893</lpage><pub-id pub-id-type="doi">10.1016/j.cell.2018.03.041</pub-id><pub-id pub-id-type="pmid">29681456</pub-id></element-citation></ref><ref id="bib42"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Knouse</surname><given-names>KA</given-names></name><name><surname>Davoli</surname><given-names>T</given-names></name><name><surname>Elledge</surname><given-names>SJ</given-names></name><name><surname>Amon</surname><given-names>A</given-names></name></person-group><year iso-8601-date="2017">2017</year><article-title>Aneuploidy in Cancer: Seq-ing Answers to Old Questions</article-title><source>Annual Review of Cancer Biology</source><volume>1</volume><fpage>335</fpage><lpage>354</lpage><pub-id pub-id-type="doi">10.1146/annurev-cancerbio-042616-072231</pub-id></element-citation></ref><ref id="bib43"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Laughney</surname><given-names>AM</given-names></name><name><surname>Elizalde</surname><given-names>S</given-names></name><name><surname>Genovese</surname><given-names>G</given-names></name><name><surname>Bakhoum</surname><given-names>SF</given-names></name></person-group><year iso-8601-date="2015">2015</year><article-title>Dynamics of Tumor Heterogeneity Derived from Clonal Karyotypic Evolution</article-title><source>Cell Reports</source><volume>12</volume><fpage>809</fpage><lpage>820</lpage><pub-id pub-id-type="doi">10.1016/j.celrep.2015.06.065</pub-id><pub-id pub-id-type="pmid">26212324</pub-id></element-citation></ref><ref id="bib44"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Lee</surname><given-names>AJX</given-names></name><name><surname>Endesfelder</surname><given-names>D</given-names></name><name><surname>Rowan</surname><given-names>AJ</given-names></name><name><surname>Walther</surname><given-names>A</given-names></name><name><surname>Birkbak</surname><given-names>NJ</given-names></name><name><surname>Futreal</surname><given-names>PA</given-names></name><name><surname>Downward</surname><given-names>J</given-names></name><name><surname>Szallasi</surname><given-names>Z</given-names></name><name><surname>Tomlinson</surname><given-names>IPM</given-names></name><name><surname>Howell</surname><given-names>M</given-names></name><name><surname>Kschischo</surname><given-names>M</given-names></name><name><surname>Swanton</surname><given-names>C</given-names></name></person-group><year iso-8601-date="2011">2011</year><article-title>Chromosomal instability confers intrinsic multidrug resistance</article-title><source>Cancer Research</source><volume>71</volume><fpage>1858</fpage><lpage>1870</lpage><pub-id pub-id-type="doi">10.1158/0008-5472.CAN-10-3604</pub-id><pub-id pub-id-type="pmid">21363922</pub-id></element-citation></ref><ref id="bib45"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Leung</surname><given-names>ML</given-names></name><name><surname>Wang</surname><given-names>Y</given-names></name><name><surname>Kim</surname><given-names>C</given-names></name><name><surname>Gao</surname><given-names>R</given-names></name><name><surname>Jiang</surname><given-names>J</given-names></name><name><surname>Sei</surname><given-names>E</given-names></name><name><surname>Navin</surname><given-names>NE</given-names></name></person-group><year iso-8601-date="2016">2016</year><article-title>Highly multiplexed targeted DNA sequencing from single nuclei</article-title><source>Nature Protocols</source><volume>11</volume><fpage>214</fpage><lpage>235</lpage><pub-id pub-id-type="doi">10.1038/nprot.2016.005</pub-id><pub-id pub-id-type="pmid">26741407</pub-id></element-citation></ref><ref id="bib46"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Levine</surname><given-names>MS</given-names></name><name><surname>Bakker</surname><given-names>B</given-names></name><name><surname>Boeckx</surname><given-names>B</given-names></name><name><surname>Moyett</surname><given-names>J</given-names></name><name><surname>Lu</surname><given-names>J</given-names></name><name><surname>Vitre</surname><given-names>B</given-names></name><name><surname>Spierings</surname><given-names>DC</given-names></name><name><surname>Lansdorp</surname><given-names>PM</given-names></name><name><surname>Cleveland</surname><given-names>DW</given-names></name><name><surname>Lambrechts</surname><given-names>D</given-names></name><name><surname>Foijer</surname><given-names>F</given-names></name><name><surname>Holland</surname><given-names>AJ</given-names></name></person-group><year iso-8601-date="2017">2017</year><article-title>Centrosome Amplification Is Sufficient to Promote Spontaneous Tumorigenesis in Mammals</article-title><source>Developmental Cell</source><volume>40</volume><fpage>313</fpage><lpage>322</lpage><pub-id pub-id-type="doi">10.1016/j.devcel.2016.12.022</pub-id><pub-id pub-id-type="pmid">28132847</pub-id></element-citation></ref><ref id="bib47"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>López</surname><given-names>S</given-names></name><name><surname>Lim</surname><given-names>EL</given-names></name><name><surname>Horswell</surname><given-names>S</given-names></name><name><surname>Haase</surname><given-names>K</given-names></name><name><surname>Huebner</surname><given-names>A</given-names></name><name><surname>Dietzen</surname><given-names>M</given-names></name><name><surname>Mourikis</surname><given-names>TP</given-names></name><name><surname>Watkins</surname><given-names>TBK</given-names></name><name><surname>Rowan</surname><given-names>A</given-names></name><name><surname>Dewhurst</surname><given-names>SM</given-names></name><name><surname>Birkbak</surname><given-names>NJ</given-names></name><name><surname>Wilson</surname><given-names>GA</given-names></name><name><surname>Van Loo</surname><given-names>P</given-names></name><name><surname>Jamal-Hanjani</surname><given-names>M</given-names></name><name><surname>Swanton</surname><given-names>C</given-names></name><name><surname>McGranahan</surname><given-names>N</given-names></name><collab>TRACERx Consortium</collab></person-group><year iso-8601-date="2020">2020</year><article-title>Interplay between whole-genome doubling and the accumulation of deleterious alterations in cancer evolution</article-title><source>Nature Genetics</source><volume>52</volume><fpage>283</fpage><lpage>293</lpage><pub-id pub-id-type="doi">10.1038/s41588-020-0584-7</pub-id><pub-id pub-id-type="pmid">32139907</pub-id></element-citation></ref><ref id="bib48"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>López-García</surname><given-names>C</given-names></name><name><surname>Sansregret</surname><given-names>L</given-names></name><name><surname>Domingo</surname><given-names>E</given-names></name><name><surname>McGranahan</surname><given-names>N</given-names></name><name><surname>Hobor</surname><given-names>S</given-names></name><name><surname>Birkbak</surname><given-names>NJ</given-names></name><name><surname>Horswell</surname><given-names>S</given-names></name><name><surname>Grönroos</surname><given-names>E</given-names></name><name><surname>Favero</surname><given-names>F</given-names></name><name><surname>Rowan</surname><given-names>AJ</given-names></name><name><surname>Matthews</surname><given-names>N</given-names></name><name><surname>Begum</surname><given-names>S</given-names></name><name><surname>Phillimore</surname><given-names>B</given-names></name><name><surname>Burrell</surname><given-names>R</given-names></name><name><surname>Oukrif</surname><given-names>D</given-names></name><name><surname>Spencer-Dene</surname><given-names>B</given-names></name><name><surname>Kovac</surname><given-names>M</given-names></name><name><surname>Stamp</surname><given-names>G</given-names></name><name><surname>Stewart</surname><given-names>A</given-names></name><name><surname>Danielsen</surname><given-names>H</given-names></name><name><surname>Novelli</surname><given-names>M</given-names></name><name><surname>Tomlinson</surname><given-names>I</given-names></name><name><surname>Swanton</surname><given-names>C</given-names></name></person-group><year iso-8601-date="2017">2017</year><article-title>BCL9L Dysfunction Impairs Caspase-2 Expression Permitting Aneuploidy Tolerance in Colorectal Cancer</article-title><source>Cancer Cell</source><volume>31</volume><fpage>79</fpage><lpage>93</lpage><pub-id pub-id-type="doi">10.1016/j.ccell.2016.11.001</pub-id><pub-id pub-id-type="pmid">28073006</pub-id></element-citation></ref><ref id="bib49"><element-citation publication-type="preprint"><person-group person-group-type="author"><name><surname>Lukow</surname><given-names>DA</given-names></name><name><surname>Sausville</surname><given-names>EL</given-names></name><name><surname>Suri</surname><given-names>P</given-names></name><name><surname>Chunduri</surname><given-names>NK</given-names></name><name><surname>Leu</surname><given-names>J</given-names></name><name><surname>Kendall</surname><given-names>J</given-names></name><name><surname>Wang</surname><given-names>Z</given-names></name><name><surname>Storchova</surname><given-names>Z</given-names></name><name><surname>Sheltzer</surname><given-names>JM</given-names></name></person-group><year iso-8601-date="2020">2020</year><article-title>Chromosomal Instability Accelerates the Evolution of Resistance to Anti-Cancer Therapies</article-title><source>bioRxiv</source><pub-id pub-id-type="doi">10.1101/2020.09.25.314229</pub-id></element-citation></ref><ref id="bib50"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Manceau</surname><given-names>M</given-names></name><name><surname>Lambert</surname><given-names>A</given-names></name><name><surname>Morlon</surname><given-names>H</given-names></name></person-group><year iso-8601-date="2015">2015</year><article-title>Phylogenies support out-of-equilibrium models of biodiversity</article-title><source>Ecology Letters</source><volume>18</volume><fpage>347</fpage><lpage>356</lpage><pub-id pub-id-type="doi">10.1111/ele.12415</pub-id><pub-id pub-id-type="pmid">25711418</pub-id></element-citation></ref><ref id="bib51"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Mooers</surname><given-names>AO</given-names></name><name><surname>Heard</surname><given-names>SB</given-names></name></person-group><year iso-8601-date="1997">1997</year><article-title>Inferring Evolutionary Process from Phylogenetic Tree Shape</article-title><source>The Quarterly Review of Biology</source><volume>72</volume><fpage>31</fpage><lpage>54</lpage><pub-id pub-id-type="doi">10.1086/419657</pub-id></element-citation></ref><ref id="bib52"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Navin</surname><given-names>N</given-names></name><name><surname>Kendall</surname><given-names>J</given-names></name><name><surname>Troge</surname><given-names>J</given-names></name><name><surname>Andrews</surname><given-names>P</given-names></name><name><surname>Rodgers</surname><given-names>L</given-names></name><name><surname>McIndoo</surname><given-names>J</given-names></name><name><surname>Cook</surname><given-names>K</given-names></name><name><surname>Stepansky</surname><given-names>A</given-names></name><name><surname>Levy</surname><given-names>D</given-names></name><name><surname>Esposito</surname><given-names>D</given-names></name><name><surname>Muthuswamy</surname><given-names>L</given-names></name><name><surname>Krasnitz</surname><given-names>A</given-names></name><name><surname>McCombie</surname><given-names>WR</given-names></name><name><surname>Hicks</surname><given-names>J</given-names></name><name><surname>Wigler</surname><given-names>M</given-names></name></person-group><year iso-8601-date="2011">2011</year><article-title>Tumour evolution inferred by single-cell sequencing</article-title><source>Nature</source><volume>472</volume><fpage>90</fpage><lpage>94</lpage><pub-id pub-id-type="doi">10.1038/nature09807</pub-id><pub-id pub-id-type="pmid">21399628</pub-id></element-citation></ref><ref id="bib53"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Neher</surname><given-names>RA</given-names></name><name><surname>Russell</surname><given-names>CA</given-names></name><name><surname>Shraiman</surname><given-names>BI</given-names></name></person-group><year iso-8601-date="2014">2014</year><article-title>Predicting evolution from the shape of genealogical trees</article-title><source>eLife</source><volume>3</volume><elocation-id>e03568</elocation-id><pub-id pub-id-type="doi">10.7554/eLife.03568</pub-id><pub-id pub-id-type="pmid">25385532</pub-id></element-citation></ref><ref id="bib54"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Nelson</surname><given-names>L</given-names></name><name><surname>Tighe</surname><given-names>A</given-names></name><name><surname>Golder</surname><given-names>A</given-names></name><name><surname>Littler</surname><given-names>S</given-names></name><name><surname>Bakker</surname><given-names>B</given-names></name><name><surname>Moralli</surname><given-names>D</given-names></name><name><surname>Murtuza Baker</surname><given-names>S</given-names></name><name><surname>Donaldson</surname><given-names>IJ</given-names></name><name><surname>Spierings</surname><given-names>DCJ</given-names></name><name><surname>Wardenaar</surname><given-names>R</given-names></name><name><surname>Neale</surname><given-names>B</given-names></name><name><surname>Burghel</surname><given-names>GJ</given-names></name><name><surname>Winter-Roach</surname><given-names>B</given-names></name><name><surname>Edmondson</surname><given-names>R</given-names></name><name><surname>Clamp</surname><given-names>AR</given-names></name><name><surname>Jayson</surname><given-names>GC</given-names></name><name><surname>Desai</surname><given-names>S</given-names></name><name><surname>Green</surname><given-names>CM</given-names></name><name><surname>Hayes</surname><given-names>A</given-names></name><name><surname>Foijer</surname><given-names>F</given-names></name><name><surname>Morgan</surname><given-names>RD</given-names></name><name><surname>Taylor</surname><given-names>SS</given-names></name></person-group><year iso-8601-date="2020">2020</year><article-title>A living biobank of ovarian cancer ex vivo models reveals profound mitotic heterogeneity</article-title><source>Nature Communications</source><volume>11</volume><elocation-id>822</elocation-id><pub-id pub-id-type="doi">10.1038/s41467-020-14551-2</pub-id><pub-id pub-id-type="pmid">32054838</pub-id></element-citation></ref><ref id="bib55"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Nicholson</surname><given-names>JM</given-names></name><name><surname>Macedo</surname><given-names>JC</given-names></name><name><surname>Mattingly</surname><given-names>AJ</given-names></name><name><surname>Wangsa</surname><given-names>D</given-names></name><name><surname>Camps</surname><given-names>J</given-names></name><name><surname>Lima</surname><given-names>V</given-names></name><name><surname>Gomes</surname><given-names>AM</given-names></name><name><surname>Dória</surname><given-names>S</given-names></name><name><surname>Ried</surname><given-names>T</given-names></name><name><surname>Logarinho</surname><given-names>E</given-names></name><name><surname>Cimini</surname><given-names>D</given-names></name></person-group><year iso-8601-date="2015">2015</year><article-title>Chromosome mis-segregation and cytokinesis failure in trisomic human cells</article-title><source>eLife</source><volume>4</volume><elocation-id>e05068</elocation-id><pub-id pub-id-type="doi">10.7554/eLife.05068</pub-id><pub-id pub-id-type="pmid">25942454</pub-id></element-citation></ref><ref id="bib56"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Nowak</surname><given-names>MA</given-names></name><name><surname>Komarova</surname><given-names>NL</given-names></name><name><surname>Sengupta</surname><given-names>A</given-names></name><name><surname>Jallepalli</surname><given-names>PV</given-names></name><name><surname>Shih</surname><given-names>IM</given-names></name><name><surname>Vogelstein</surname><given-names>B</given-names></name><name><surname>Lengauer</surname><given-names>C</given-names></name></person-group><year iso-8601-date="2002">2002</year><article-title>The role of chromosomal instability in tumor initiation</article-title><source>PNAS</source><volume>99</volume><fpage>16226</fpage><lpage>16231</lpage><pub-id pub-id-type="doi">10.1073/pnas.202617399</pub-id><pub-id pub-id-type="pmid">12446840</pub-id></element-citation></ref><ref id="bib57"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Orr</surname><given-names>B</given-names></name><name><surname>Talje</surname><given-names>L</given-names></name><name><surname>Liu</surname><given-names>Z</given-names></name><name><surname>Kwok</surname><given-names>BH</given-names></name><name><surname>Compton</surname><given-names>DA</given-names></name></person-group><year iso-8601-date="2016">2016</year><article-title>Adaptive Resistance to an Inhibitor of Chromosomal Instability in Human Cancer Cells</article-title><source>Cell Reports</source><volume>17</volume><fpage>1755</fpage><lpage>1763</lpage><pub-id pub-id-type="doi">10.1016/j.celrep.2016.10.030</pub-id><pub-id pub-id-type="pmid">27829147</pub-id></element-citation></ref><ref id="bib58"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Pavelka</surname><given-names>N</given-names></name><name><surname>Rancati</surname><given-names>G</given-names></name><name><surname>Zhu</surname><given-names>J</given-names></name><name><surname>Bradford</surname><given-names>WD</given-names></name><name><surname>Saraf</surname><given-names>A</given-names></name><name><surname>Florens</surname><given-names>L</given-names></name><name><surname>Sanderson</surname><given-names>BW</given-names></name><name><surname>Hattem</surname><given-names>GL</given-names></name><name><surname>Li</surname><given-names>R</given-names></name></person-group><year iso-8601-date="2010">2010</year><article-title>Aneuploidy confers quantitative proteome changes and phenotypic variation in budding yeast</article-title><source>Nature</source><volume>468</volume><fpage>321</fpage><lpage>325</lpage><pub-id pub-id-type="doi">10.1038/nature09529</pub-id><pub-id pub-id-type="pmid">20962780</pub-id></element-citation></ref><ref id="bib59"><element-citation publication-type="software"><person-group person-group-type="author"><collab>R Development Core Team</collab></person-group><year iso-8601-date="2021">2021</year><data-title>R: A language and environment for statistical computing</data-title><version designator="2.6.2">2.6.2</version><publisher-loc>Vienna, Austria</publisher-loc><publisher-name>R Foundation for Statistical Computing</publisher-name><ext-link ext-link-type="uri" xlink:href="https://www.R-project.org/">https://www.R-project.org/</ext-link></element-citation></ref><ref id="bib60"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Ravichandran</surname><given-names>MC</given-names></name><name><surname>Fink</surname><given-names>S</given-names></name><name><surname>Clarke</surname><given-names>MN</given-names></name><name><surname>Hofer</surname><given-names>FC</given-names></name><name><surname>Campbell</surname><given-names>CS</given-names></name></person-group><year iso-8601-date="2018">2018</year><article-title>Genetic interactions between specific chromosome copy number alterations dictate complex aneuploidy patterns</article-title><source>Genes & Development</source><volume>32</volume><fpage>1485</fpage><lpage>1498</lpage><pub-id pub-id-type="doi">10.1101/gad.319400.118</pub-id><pub-id pub-id-type="pmid">30463904</pub-id></element-citation></ref><ref id="bib61"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Sack</surname><given-names>LM</given-names></name><name><surname>Davoli</surname><given-names>T</given-names></name><name><surname>Li</surname><given-names>MZ</given-names></name><name><surname>Li</surname><given-names>Y</given-names></name><name><surname>Xu</surname><given-names>Q</given-names></name><name><surname>Naxerova</surname><given-names>K</given-names></name><name><surname>Wooten</surname><given-names>EC</given-names></name><name><surname>Bernardi</surname><given-names>RJ</given-names></name><name><surname>Martin</surname><given-names>TD</given-names></name><name><surname>Chen</surname><given-names>T</given-names></name><name><surname>Leng</surname><given-names>Y</given-names></name><name><surname>Liang</surname><given-names>AC</given-names></name><name><surname>Scorsone</surname><given-names>KA</given-names></name><name><surname>Westbrook</surname><given-names>TF</given-names></name><name><surname>Wong</surname><given-names>K-K</given-names></name><name><surname>Elledge</surname><given-names>SJ</given-names></name></person-group><year iso-8601-date="2018">2018</year><article-title>Profound Tissue Specificity in Proliferation Control Underlies Cancer Drivers and Aneuploidy Patterns</article-title><source>Cell</source><volume>173</volume><fpage>499</fpage><lpage>514</lpage><pub-id pub-id-type="doi">10.1016/j.cell.2018.02.037</pub-id><pub-id pub-id-type="pmid">29576454</pub-id></element-citation></ref><ref id="bib62"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Santaguida</surname><given-names>S</given-names></name><name><surname>Richardson</surname><given-names>A</given-names></name><name><surname>Iyer</surname><given-names>DR</given-names></name><name><surname>M’Saad</surname><given-names>O</given-names></name><name><surname>Zasadil</surname><given-names>L</given-names></name><name><surname>Knouse</surname><given-names>KA</given-names></name><name><surname>Wong</surname><given-names>YL</given-names></name><name><surname>Rhind</surname><given-names>N</given-names></name><name><surname>Desai</surname><given-names>A</given-names></name><name><surname>Amon</surname><given-names>A</given-names></name></person-group><year iso-8601-date="2017">2017</year><article-title>Chromosome Mis-segregation Generates Cell-Cycle-Arrested Cells with Complex Karyotypes that Are Eliminated by the Immune System</article-title><source>Developmental Cell</source><volume>41</volume><fpage>638</fpage><lpage>651</lpage><pub-id pub-id-type="doi">10.1016/j.devcel.2017.05.022</pub-id><pub-id pub-id-type="pmid">28633018</pub-id></element-citation></ref><ref id="bib63"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Scott</surname><given-names>JG</given-names></name><name><surname>Maini</surname><given-names>PK</given-names></name><name><surname>Anderson</surname><given-names>ARA</given-names></name><name><surname>Fletcher</surname><given-names>AG</given-names></name></person-group><year iso-8601-date="2020">2020</year><article-title>Inferring Tumor Proliferative Organization from Phylogenetic Tree Measures in a Computational Model</article-title><source>Systematic Biology</source><volume>69</volume><fpage>623</fpage><lpage>637</lpage><pub-id pub-id-type="doi">10.1093/sysbio/syz070</pub-id><pub-id pub-id-type="pmid">31665523</pub-id></element-citation></ref><ref id="bib64"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Scribano</surname><given-names>CM</given-names></name><name><surname>Wan</surname><given-names>J</given-names></name><name><surname>Esbona</surname><given-names>K</given-names></name><name><surname>Tucker</surname><given-names>JB</given-names></name><name><surname>Lasek</surname><given-names>A</given-names></name><name><surname>Zhou</surname><given-names>AS</given-names></name><name><surname>Zasadil</surname><given-names>LM</given-names></name><name><surname>Molini</surname><given-names>R</given-names></name><name><surname>Fitzgerald</surname><given-names>J</given-names></name><name><surname>Lager</surname><given-names>AM</given-names></name><name><surname>Laffin</surname><given-names>JJ</given-names></name><name><surname>Correia-Staudt</surname><given-names>K</given-names></name><name><surname>Wisinski</surname><given-names>KB</given-names></name><name><surname>Tevaarwerk</surname><given-names>AJ</given-names></name><name><surname>O’Regan</surname><given-names>R</given-names></name><name><surname>McGregor</surname><given-names>SM</given-names></name><name><surname>Fowler</surname><given-names>AM</given-names></name><name><surname>Chappell</surname><given-names>RJ</given-names></name><name><surname>Bugni</surname><given-names>TS</given-names></name><name><surname>Burkard</surname><given-names>ME</given-names></name><name><surname>Weaver</surname><given-names>BA</given-names></name></person-group><year iso-8601-date="2021">2021</year><article-title>Chromosomal instability sensitizes patient breast tumors to multipolar divisions induced by paclitaxel</article-title><source>Science Translational Medicine</source><volume>13</volume><elocation-id>eabd4811</elocation-id><pub-id pub-id-type="doi">10.1126/scitranslmed.abd4811</pub-id><pub-id pub-id-type="pmid">34516829</pub-id></element-citation></ref><ref id="bib65"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Sheltzer</surname><given-names>JM</given-names></name><name><surname>Amon</surname><given-names>A</given-names></name></person-group><year iso-8601-date="2011">2011</year><article-title>The aneuploidy paradox: costs and benefits of an incorrect karyotype</article-title><source>Trends in Genetics</source><volume>27</volume><fpage>446</fpage><lpage>453</lpage><pub-id pub-id-type="doi">10.1016/j.tig.2011.07.003</pub-id><pub-id pub-id-type="pmid">21872963</pub-id></element-citation></ref><ref id="bib66"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Sheltzer</surname><given-names>JM</given-names></name></person-group><year iso-8601-date="2013">2013</year><article-title>A Transcriptional and Metabolic Signature of Primary Aneuploidy Is Present in Chromosomally Unstable Cancer Cells and Informs Clinical Prognosis</article-title><source>Cancer Research</source><volume>73</volume><fpage>6401</fpage><lpage>6412</lpage><pub-id pub-id-type="doi">10.1158/0008-5472.CAN-13-0749</pub-id></element-citation></ref><ref id="bib67"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Sheltzer</surname><given-names>JM</given-names></name><name><surname>Ko</surname><given-names>JH</given-names></name><name><surname>Replogle</surname><given-names>JM</given-names></name><name><surname>Habibe Burgos</surname><given-names>NC</given-names></name><name><surname>Chung</surname><given-names>ES</given-names></name><name><surname>Meehl</surname><given-names>CM</given-names></name><name><surname>Sayles</surname><given-names>NM</given-names></name><name><surname>Passerini</surname><given-names>V</given-names></name><name><surname>Storchova</surname><given-names>Z</given-names></name><name><surname>Amon</surname><given-names>A</given-names></name></person-group><year iso-8601-date="2017">2017</year><article-title>Single-chromosome Gains Commonly Function as Tumor Suppressors</article-title><source>Cancer Cell</source><volume>31</volume><fpage>240</fpage><lpage>255</lpage><pub-id pub-id-type="doi">10.1016/j.ccell.2016.12.004</pub-id><pub-id pub-id-type="pmid">28089890</pub-id></element-citation></ref><ref id="bib68"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Silk</surname><given-names>AD</given-names></name><name><surname>Zasadil</surname><given-names>LM</given-names></name><name><surname>Holland</surname><given-names>AJ</given-names></name><name><surname>Vitre</surname><given-names>B</given-names></name><name><surname>Cleveland</surname><given-names>DW</given-names></name><name><surname>Weaver</surname><given-names>BA</given-names></name></person-group><year iso-8601-date="2013">2013</year><article-title>Chromosome missegregation rate predicts whether aneuploidy will promote or suppress tumors</article-title><source>PNAS</source><volume>110</volume><fpage>E4134</fpage><lpage>E4141</lpage><pub-id pub-id-type="doi">10.1073/pnas.1317042110</pub-id><pub-id pub-id-type="pmid">24133140</pub-id></element-citation></ref><ref id="bib69"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Simões-Sousa</surname><given-names>S</given-names></name><name><surname>Littler</surname><given-names>S</given-names></name><name><surname>Thompson</surname><given-names>SL</given-names></name><name><surname>Minshall</surname><given-names>P</given-names></name><name><surname>Whalley</surname><given-names>H</given-names></name><name><surname>Bakker</surname><given-names>B</given-names></name><name><surname>Belkot</surname><given-names>K</given-names></name><name><surname>Moralli</surname><given-names>D</given-names></name><name><surname>Bronder</surname><given-names>D</given-names></name><name><surname>Tighe</surname><given-names>A</given-names></name><name><surname>Spierings</surname><given-names>DCJ</given-names></name><name><surname>Bah</surname><given-names>N</given-names></name><name><surname>Graham</surname><given-names>J</given-names></name><name><surname>Nelson</surname><given-names>L</given-names></name><name><surname>Green</surname><given-names>CM</given-names></name><name><surname>Foijer</surname><given-names>F</given-names></name><name><surname>Townsend</surname><given-names>PA</given-names></name><name><surname>Taylor</surname><given-names>SS</given-names></name></person-group><year iso-8601-date="2018">2018</year><article-title>The p38α Stress Kinase Suppresses Aneuploidy Tolerance by Inhibiting Hif-1α</article-title><source>Cell Reports</source><volume>25</volume><fpage>749</fpage><lpage>760</lpage><pub-id pub-id-type="doi">10.1016/j.celrep.2018.09.060</pub-id><pub-id pub-id-type="pmid">30332653</pub-id></element-citation></ref><ref id="bib70"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Soto</surname><given-names>M</given-names></name><name><surname>Raaijmakers</surname><given-names>JA</given-names></name><name><surname>Bakker</surname><given-names>B</given-names></name><name><surname>Spierings</surname><given-names>DCJ</given-names></name><name><surname>Lansdorp</surname><given-names>PM</given-names></name><name><surname>Foijer</surname><given-names>F</given-names></name><name><surname>Medema</surname><given-names>RH</given-names></name></person-group><year iso-8601-date="2017">2017</year><article-title>p53 Prohibits Propagation of Chromosome Segregation Errors that Produce Structural Aneuploidies</article-title><source>Cell Reports</source><volume>19</volume><fpage>2423</fpage><lpage>2431</lpage><pub-id pub-id-type="doi">10.1016/j.celrep.2017.05.055</pub-id><pub-id pub-id-type="pmid">28636931</pub-id></element-citation></ref><ref id="bib71"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Sottoriva</surname><given-names>A</given-names></name><name><surname>Kang</surname><given-names>H</given-names></name><name><surname>Ma</surname><given-names>Z</given-names></name><name><surname>Graham</surname><given-names>TA</given-names></name><name><surname>Salomon</surname><given-names>MP</given-names></name><name><surname>Zhao</surname><given-names>J</given-names></name><name><surname>Marjoram</surname><given-names>P</given-names></name><name><surname>Siegmund</surname><given-names>K</given-names></name><name><surname>Press</surname><given-names>MF</given-names></name><name><surname>Shibata</surname><given-names>D</given-names></name><name><surname>Curtis</surname><given-names>C</given-names></name></person-group><year iso-8601-date="2015">2015</year><article-title>A Big Bang model of human colorectal tumor growth</article-title><source>Nature Genetics</source><volume>47</volume><fpage>209</fpage><lpage>216</lpage><pub-id pub-id-type="doi">10.1038/ng.3214</pub-id><pub-id pub-id-type="pmid">25665006</pub-id></element-citation></ref><ref id="bib72"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Starostik</surname><given-names>MR</given-names></name><name><surname>Sosina</surname><given-names>OA</given-names></name><name><surname>McCoy</surname><given-names>RC</given-names></name></person-group><year iso-8601-date="2020">2020</year><article-title>Single-cell analysis of human embryos reveals diverse patterns of aneuploidy and mosaicism</article-title><source>Genome Research</source><volume>30</volume><fpage>814</fpage><lpage>825</lpage><pub-id pub-id-type="doi">10.1101/gr.262774.120</pub-id><pub-id pub-id-type="pmid">32641298</pub-id></element-citation></ref><ref id="bib73"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Swanton</surname><given-names>C</given-names></name><name><surname>Nicke</surname><given-names>B</given-names></name><name><surname>Schuett</surname><given-names>M</given-names></name><name><surname>Eklund</surname><given-names>AC</given-names></name><name><surname>Ng</surname><given-names>C</given-names></name><name><surname>Li</surname><given-names>Q</given-names></name><name><surname>Hardcastle</surname><given-names>T</given-names></name><name><surname>Lee</surname><given-names>A</given-names></name><name><surname>Roy</surname><given-names>R</given-names></name><name><surname>East</surname><given-names>P</given-names></name><name><surname>Kschischo</surname><given-names>M</given-names></name><name><surname>Endesfelder</surname><given-names>D</given-names></name><name><surname>Wylie</surname><given-names>P</given-names></name><name><surname>Kim</surname><given-names>SN</given-names></name><name><surname>Chen</surname><given-names>JG</given-names></name><name><surname>Howell</surname><given-names>M</given-names></name><name><surname>Ried</surname><given-names>T</given-names></name><name><surname>Habermann</surname><given-names>JK</given-names></name><name><surname>Auer</surname><given-names>G</given-names></name><name><surname>Brenton</surname><given-names>JD</given-names></name><name><surname>Szallasi</surname><given-names>Z</given-names></name><name><surname>Downward</surname><given-names>J</given-names></name></person-group><year iso-8601-date="2009">2009</year><article-title>Chromosomal instability determines taxane response</article-title><source>PNAS</source><volume>106</volume><fpage>8671</fpage><lpage>8676</lpage><pub-id pub-id-type="doi">10.1073/pnas.0811835106</pub-id><pub-id pub-id-type="pmid">19458043</pub-id></element-citation></ref><ref id="bib74"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Thompson</surname><given-names>SL</given-names></name><name><surname>Compton</surname><given-names>DA</given-names></name></person-group><year iso-8601-date="2008">2008</year><article-title>Examining the link between chromosomal instability and aneuploidy in human cells</article-title><source>The Journal of Cell Biology</source><volume>180</volume><fpage>665</fpage><lpage>672</lpage><pub-id pub-id-type="doi">10.1083/jcb.200712029</pub-id><pub-id pub-id-type="pmid">18283116</pub-id></element-citation></ref><ref id="bib75"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Vasudevan</surname><given-names>A</given-names></name><name><surname>Baruah</surname><given-names>PS</given-names></name><name><surname>Smith</surname><given-names>JC</given-names></name><name><surname>Wang</surname><given-names>Z</given-names></name><name><surname>Sayles</surname><given-names>NM</given-names></name><name><surname>Andrews</surname><given-names>P</given-names></name><name><surname>Kendall</surname><given-names>J</given-names></name><name><surname>Leu</surname><given-names>J</given-names></name><name><surname>Chunduri</surname><given-names>NK</given-names></name><name><surname>Levy</surname><given-names>D</given-names></name><name><surname>Wigler</surname><given-names>M</given-names></name><name><surname>Storchová</surname><given-names>Z</given-names></name><name><surname>Sheltzer</surname><given-names>JM</given-names></name></person-group><year iso-8601-date="2020">2020</year><article-title>Single-Chromosomal Gains Can Function as Metastasis Suppressors and Promoters in Colon Cancer</article-title><source>Developmental Cell</source><volume>52</volume><fpage>413</fpage><lpage>428</lpage><pub-id pub-id-type="doi">10.1016/j.devcel.2020.01.034</pub-id><pub-id pub-id-type="pmid">32097652</pub-id></element-citation></ref><ref id="bib76"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Wang</surname><given-names>Y</given-names></name><name><surname>Waters</surname><given-names>J</given-names></name><name><surname>Leung</surname><given-names>ML</given-names></name><name><surname>Unruh</surname><given-names>A</given-names></name><name><surname>Roh</surname><given-names>W</given-names></name><name><surname>Shi</surname><given-names>X</given-names></name><name><surname>Chen</surname><given-names>K</given-names></name><name><surname>Scheet</surname><given-names>P</given-names></name><name><surname>Vattathil</surname><given-names>S</given-names></name><name><surname>Liang</surname><given-names>H</given-names></name><name><surname>Multani</surname><given-names>A</given-names></name><name><surname>Zhang</surname><given-names>H</given-names></name><name><surname>Zhao</surname><given-names>R</given-names></name><name><surname>Michor</surname><given-names>F</given-names></name><name><surname>Meric-Bernstam</surname><given-names>F</given-names></name><name><surname>Navin</surname><given-names>NE</given-names></name></person-group><year iso-8601-date="2014">2014</year><article-title>Clonal evolution in breast cancer revealed by single nucleus genome sequencing</article-title><source>Nature</source><volume>512</volume><fpage>155</fpage><lpage>160</lpage><pub-id pub-id-type="doi">10.1038/nature13600</pub-id><pub-id pub-id-type="pmid">25079324</pub-id></element-citation></ref><ref id="bib77"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Weaver</surname><given-names>BAA</given-names></name><name><surname>Cleveland</surname><given-names>DW</given-names></name></person-group><year iso-8601-date="2006">2006</year><article-title>Does aneuploidy cause cancer?</article-title><source>Current Opinion in Cell Biology</source><volume>18</volume><fpage>658</fpage><lpage>667</lpage><pub-id pub-id-type="doi">10.1016/j.ceb.2006.10.002</pub-id><pub-id pub-id-type="pmid">17046232</pub-id></element-citation></ref><ref id="bib78"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Weaver</surname><given-names>BAA</given-names></name><name><surname>Silk</surname><given-names>AD</given-names></name><name><surname>Montagna</surname><given-names>C</given-names></name><name><surname>Verdier-Pinard</surname><given-names>P</given-names></name><name><surname>Cleveland</surname><given-names>DW</given-names></name></person-group><year iso-8601-date="2007">2007</year><article-title>Aneuploidy acts both oncogenically and as a tumor suppressor</article-title><source>Cancer Cell</source><volume>11</volume><fpage>25</fpage><lpage>36</lpage><pub-id pub-id-type="doi">10.1016/j.ccr.2006.12.003</pub-id><pub-id pub-id-type="pmid">17189716</pub-id></element-citation></ref><ref id="bib79"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Weaver</surname><given-names>BA</given-names></name><name><surname>Cleveland</surname><given-names>DW</given-names></name></person-group><year iso-8601-date="2008">2008</year><article-title>The aneuploidy paradox in cell growth and tumorigenesis</article-title><source>Cancer Cell</source><volume>14</volume><fpage>431</fpage><lpage>433</lpage><pub-id pub-id-type="doi">10.1016/j.ccr.2008.11.011</pub-id><pub-id pub-id-type="pmid">19061834</pub-id></element-citation></ref><ref id="bib80"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Weaver</surname><given-names>BAA</given-names></name><name><surname>Cleveland</surname><given-names>DW</given-names></name></person-group><year iso-8601-date="2009">2009</year><article-title>The role of aneuploidy in promoting and suppressing tumors</article-title><source>The Journal of Cell Biology</source><volume>185</volume><fpage>935</fpage><lpage>937</lpage><pub-id pub-id-type="doi">10.1083/jcb.200905098</pub-id><pub-id pub-id-type="pmid">19528293</pub-id></element-citation></ref><ref id="bib81"><element-citation publication-type="web"><person-group person-group-type="author"><name><surname>Wilensky</surname><given-names>U</given-names></name></person-group><year iso-8601-date="1999">1999</year><article-title>NetLogo</article-title><ext-link ext-link-type="uri" xlink:href="http://ccl.northwestern.edu/netlogo/">http://ccl.northwestern.edu/netlogo/</ext-link><date-in-citation iso-8601-date="2020-03-11">March 11, 2020</date-in-citation></element-citation></ref><ref id="bib82"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Worrall</surname><given-names>JT</given-names></name><name><surname>Tamura</surname><given-names>N</given-names></name><name><surname>Mazzagatti</surname><given-names>A</given-names></name><name><surname>Shaikh</surname><given-names>N</given-names></name><name><surname>van Lingen</surname><given-names>T</given-names></name><name><surname>Bakker</surname><given-names>B</given-names></name><name><surname>Spierings</surname><given-names>DCJ</given-names></name><name><surname>Vladimirou</surname><given-names>E</given-names></name><name><surname>Foijer</surname><given-names>F</given-names></name><name><surname>McClelland</surname><given-names>SE</given-names></name></person-group><year iso-8601-date="2018">2018</year><article-title>Non-random Mis-segregation of Human Chromosomes</article-title><source>Cell Reports</source><volume>23</volume><fpage>3366</fpage><lpage>3380</lpage><pub-id pub-id-type="doi">10.1016/j.celrep.2018.05.047</pub-id><pub-id pub-id-type="pmid">29898405</pub-id></element-citation></ref><ref id="bib83"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Zasadil</surname><given-names>LM</given-names></name><name><surname>Andersen</surname><given-names>KA</given-names></name><name><surname>Yeum</surname><given-names>D</given-names></name><name><surname>Rocque</surname><given-names>GB</given-names></name><name><surname>Wilke</surname><given-names>LG</given-names></name><name><surname>Tevaarwerk</surname><given-names>AJ</given-names></name><name><surname>Raines</surname><given-names>RT</given-names></name><name><surname>Burkard</surname><given-names>ME</given-names></name><name><surname>Weaver</surname><given-names>BA</given-names></name></person-group><year iso-8601-date="2014">2014</year><article-title>Cytotoxicity of paclitaxel in breast cancer is due to chromosome missegregation on multipolar spindles</article-title><source>Science Translational Medicine</source><volume>6</volume><elocation-id>ra43</elocation-id><pub-id pub-id-type="doi">10.1126/scitranslmed.3007965</pub-id><pub-id pub-id-type="pmid">24670687</pub-id></element-citation></ref><ref id="bib84"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Zhu</surname><given-names>J</given-names></name><name><surname>Pavelka</surname><given-names>N</given-names></name><name><surname>Bradford</surname><given-names>WD</given-names></name><name><surname>Rancati</surname><given-names>G</given-names></name><name><surname>Li</surname><given-names>R</given-names></name></person-group><year iso-8601-date="2012">2012</year><article-title>Karyotypic determinants of chromosome instability in aneuploid budding yeast</article-title><source>PLOS Genetics</source><volume>8</volume><elocation-id>e1002719</elocation-id><pub-id pub-id-type="doi">10.1371/journal.pgen.1002719</pub-id><pub-id pub-id-type="pmid">22615582</pub-id></element-citation></ref></ref-list></back><sub-article article-type="editor-report" id="sa0"><front-stub><article-id pub-id-type="doi">10.7554/eLife.69799.sa0</article-id><title-group><article-title>Editor's evaluation</article-title></title-group><contrib-group><contrib contrib-type="author"><name><surname>Marston</surname><given-names>Adèle L</given-names></name><role specific-use="editor">Reviewing Editor</role><aff><institution-wrap><institution-id institution-id-type="ror">https://ror.org/01nrxwf90</institution-id><institution>University of Edinburgh</institution></institution-wrap><country>United Kingdom</country></aff></contrib></contrib-group><related-object id="sa0ro1" link-type="continued-by" object-id="10.1101/2021.04.26.441466" object-id-type="id" xlink:href="https://sciety.org/articles/activity/10.1101/2021.04.26.441466"/></front-stub><body><p>The authors have developed a framework to quantify rates of chromosomal instability (CIN) in human tumors by fitting karyotype distributions inferred from low-depth DNA-sequencing to in silico models of CIN with karyotype selection pressures, sweeping through parameter space. This is particularly useful for the development of biomarkers for CIN, which is associated with cancer metastasis and drug resistance.</p></body></sub-article><sub-article article-type="decision-letter" id="sa1"><front-stub><article-id pub-id-type="doi">10.7554/eLife.69799.sa1</article-id><title-group><article-title>Decision letter</article-title></title-group><contrib-group content-type="section"><contrib contrib-type="editor"><name><surname>Marston</surname><given-names>Adèle L</given-names></name><role>Reviewing Editor</role><aff><institution-wrap><institution-id institution-id-type="ror">https://ror.org/01nrxwf90</institution-id><institution>University of Edinburgh</institution></institution-wrap><country>United Kingdom</country></aff></contrib></contrib-group><contrib-group><contrib contrib-type="reviewer"><name><surname>Graham</surname><given-names>Trevor A</given-names></name><role>Reviewer</role><aff><institution-wrap><institution-id institution-id-type="ror">https://ror.org/04cw6st05</institution-id><institution>Queen Mary, University of London</institution></institution-wrap><country>United Kingdom</country></aff></contrib></contrib-group></front-stub><body><boxed-text id="sa2-box1"><p>Our editorial process produces two outputs: (i) <ext-link ext-link-type="uri" xlink:href="https://sciety.org/articles/activity/10.1101/2021.04.26.441466">public reviews</ext-link> designed to be posted alongside <ext-link ext-link-type="uri" xlink:href="https://www.biorxiv.org/content/10.1101/2021.04.26.441466v1">the preprint</ext-link> for the benefit of readers; (ii) feedback on the manuscript for the authors, including requests for revisions, shown below. We also include an acceptance summary that explains what the editors found interesting or important about the work.</p></boxed-text><p><bold>Decision letter after peer review:</bold></p><p>Thank you for submitting your article "Quantifying chromosomal instability from intratumoral karyotype diversity using agent- based modeling and Bayesian inference" for consideration by <italic>eLife</italic>. Your article has been reviewed by 3 peer reviewers, and the evaluation has been overseen by a Reviewing Editor and Anna Akhmanova as the Senior Editor. The following individuals involved in review of your submission have agreed to reveal their identity: Trevor A Graham (Reviewer #3).</p><p>The reviewers have discussed their reviews with one another, and the Reviewing Editor has drafted this to help you prepare a revised submission.</p><p>Essential revisions:</p><p>1. Modelling.</p><p>Three different models to compute cell fitness based on karyotypic alteration are explored. The construction of all these models feels a little arbitrary, and the assumptions and evolutionary dynamics in each scenario should be more comprehensively explored. Specifically:</p><p>a) In the TOE model, fitness is inversely proportional to average ploidy, so it seems higher ploidies are always selected against. Is this a reasonable assumption? Why is it necessary to divide by the average ploidy?</p><p>b) In all models, is the simulated population always "out of equilibrium". If the simulations ran for longer would an "optimal karyotype" be established. Relatedly, the dynamics appear to be strongly influenced by the copy number >6 being lethal – chromosomes (in the TOE model) which are beneficial to be gained might tend to increase copy number to 5 whereas deleterious gains reduce copy number to 1 and the population rests on that "precipice". How reasonable are these "boundary conditions" and do the dynamics change significantly if they are relaxed?</p><p>c) Gains and losses appear to be treated equivalently – again is this reasonable? Especially in the TOE model where TSG gains and OG losses (and vice versa) have differing consequences. (see also point 8 below).</p><p>The modelling assumes exponential growth to a relatively small number of cells (4500) and then randomly kills half the cells to reinitiate exponential growth from 2500 cells. This regime will influence the evolutionary dynamics of the system: the random killing will cause emerging clones to often go extinct and could exacerbate the influence of drift in the system. It also effects the influence of selection, (see for example: https://www.nature.com/articles/ng.3214). Alternative growth dynamics could be implemented – such as a Wright Fisher type model of either constant or growing populations (for the construction of a growing WF see: https://pubmed.ncbi.nlm.nih.gov/25665006/) and the influence of the growth dynamics on karyotypic heterogeneity robustly assessed.</p><p>The modelling assumes only whole chromosome missegregations (and the bioinformatics data analysis averages over sub-chromsomal sized events). The authors could consider extending their analysis to handle part-chromosome events to better represent the biological data.</p><p>2. Bayesian inference.</p><p>It is a concern that unusual prior distributions are used in the ABC inference and this effects the reliability of the inference. Figure 5F and 6C show smoothed density plots for the prior distributions – which confusingly show density for S<0 – and the true priors might instead be a series of point masses at a handful of S values. This should be clarified.</p><p>It is likely that the posterior the alteration rate and S are interrelated (high S inferred when the alteration rate is high and vice versa) – so joint posteriors should be shown. Because of the interrelationship between the parameters, it is a concern that the current parameter estimates are inaccurate – currently the prior for S has zero mass for many S values, and the inferred value of the alteration rate will depend on which values of S are explored form the prior distribution. The inference should be repeated using continuous distribution over S, a uniform distribution is suggested.</p><p>When real data is analysed, only the hybrid model is compared to data, but their Figure 3 shows the diversity depends on the underlying model of selection. The authors should implement a model selection routine (one is available with the ABC-sysbio package: https://pubmed.ncbi.nlm.nih.gov/20591907/) to test which selection paradigm (if any) best represents the data. They could also consider comparing what is arguably the "null hypothesis" of neutral evolution (S=0) to a case with selection as part of this model selection, to quantitatively determine the evidence of selection in the data.</p><p>A single threshold for acceptance in the ABC algorithm (epsilon=0.05). The authors should show this is sufficiently small: a straightforward way is to plot the mean and variance of the posteriors as a function of epsilon.</p><p>Above, the possibility of temporal effects in the model are mentioned – how do temporal ("out of equilibrium" evolutionary dynamics) affect the inference?</p><p>When public data are analysed, the growth dynamics of the breast tissue, or colorectal organoid is very unlikely to match the exponential growth assumption that is assumed by the authors. These growth dynamics likely strongly determine the pattern of observed heterogeneity (exponential growth leads to large founder effects in the extant population) and so the influence of alterative growth models needs to be explored. This can also be done within a model selection framework.</p><p>The initial state of the model used to fit to public dataset is not specified and needs to be. The authors might also consider the influence of spatial sampling confounders (the breast dataset in particular in not a well-mixed population).</p><p>3. Single cell data analysis. The data are presented as if only whole chromosomal alterations occur (e.g. figure 5B). Is this actually the case? (certainly the assumption is false in breast and colon organoid datasets) Could relative read counts in say 1MB bins for each cell be provided in the appendix to reassure the reader that the types of genetic alterations occurring in their single cell data mirror the assumptions of the modelling.</p><p>As noted above, the authors should consider modelling part-chromosome alterations. They should be able to determine in their simulations how their assumption of only whole chromosome missegregation events, despite their being part-chromosomal copy number alterations in the data, affects the accuracy of the inferred chromosomal missegregation rate.</p><p>4. The presented framework is lacking expanded characterization and validation of selection models that are biologically relevant. The current framework simply applies a scalar exponent to already published fitness models for selection. It is unclear what this exponent mirrors biologically, beyond amplifying the selection pressures already explored in existing gene abundance and driver density models.</p><p>5. Related to point (4)., how is the CIN ON-OFF model in which CIN is turned off after so many cell divisions relevant biologically? Typically, CIN is a considered a trait that evolves later in cancer progression, that once tolerated, is ongoing and facilitates development of metastasis and drug resistance. A more relevant model to explore would be that of the effect of a whole genome duplication (WGD) event on population evolution, which is thought to facilitate tolerance of ensuing missegregation events (because reduce risk of nullisomy).</p><p>5. The authors utilize two models of karyotype fitness – a gene abundance model and driver density model – to evaluate impact of specific karyotypes on cellular fitness. They also include a hybrid model whose fitness effects are simply the average of these two models, which adds little value as only a weighted average. In silico results shows inferred missegregation rates are extremely disparate across the two primary models. And while a description of these differences is provided, the presented analyses do not make clear the most important question – which of these models is more clinically relevant? Toward this, in Figure 2F, the authors claim the three models approach a triploid state – which is unsupported by the in silico results. Clearly the driver model approaches a triploid state, as previously reported. But the abundance model does not and hybrid only slightly so, given that it is simply a weighted average of these two approaches. Because the authors have developed a Bayesian strategy for inferring which model parameters best fit observed data, it would be very useful to see which model best recapitulates karyotypes observed in cancer cell lines or patient materials.</p><p>6. Topological features of phylogenetic trees, while discriminatory, are largely dependent on accurate phylogenetic tree reconstruction. The latter requires more careful consideration of cell linkages beyond computing pairwise Euclidean distances and performing complete-linkage clustering. For example, a WGD event, would appear very far from its nearest cell ancestor in Euclidean space.</p><p>7. Experimental validation of the added selection exponential factor is imperative. Works have already shown models of karyotypic evolution without additional selection exponential coefficient can accurately recover rates of missegregation observed in human cell lines and cancers by fluorescent microscopy. Incorporation of this additional weight on selection pressure has not been demonstrated or validated experimentally. This would require experimental sampling of karyotypes longitudinally and is a critical piece of this manuscript's novelty.</p><p>8. It seems like this model treats chromosome gains and losses equivalently. Is this appropriate? Chromosome loss events are much more toxic than chromosome gain events – as evidenced by the fact that haploinsufficiency is widespread, and all autosomal monosomies are embryonically-lethal while many trisomies are compatible with birth and development. Can the authors consider a model in which losses exert a more significant fitness penalty that chromosome gains? (see also point 1c above).</p><p>9. Chromosomes do not missegregate at the same rate (PMID: 29898405). This point should be discussed, and, if feasible, incorporated into the authors' models.</p><p>10. Can the authors clarify their use of live cell imaging (e.g., in Figure 6G)? Certain apparent errors that are visible by live-cell imaging (like a lagging chromosome) can be resolved correctly and result in proper segregation. Is it appropriate to directly interfer missegregation rates as is done in this paper?</p><p>11. The authors should discuss in greater detail earlier mathematical models of CIN, including PMID: 26212324, 30204765, and 12446840. How does their approach improve on this prior work?</p><p>[Editors' note: further revisions were suggested prior to acceptance, as described below.]</p><p>Thank you for resubmitting your work entitled "Quantifying chromosomal instability from intratumoral karyotype diversity using agent- based modeling and Bayesian inference" for further consideration by <italic>eLife</italic>. Your revised article has been evaluated by Anna Akhmanova (Senior Editor), Reviewing Editors and two of the original reviewers.</p><p>The manuscript has been improved but there are some remaining issues that need to be addressed, as outlined below:</p><p>1. The reviewers are concerned about the choice of statistics used for the ABC analysis. While these summary statistics indeed must contain some statistical signal on the rate of chromosomal instability, the current method of "phylogenetic reconstruction" which based on euclidian distances of chromosomal copy number variation is very indirect. The reviewers are not convinced that this is an optimal way to extract relevant information from the distribution of chromosomal copy number across cells. At least they are not clearly rooted in models of clonal evolution.</p><p>Why this particular set of summary statistics was chosen, how it relates to the quantity of interest (rate of chromosomal instability), how it compares to other possible ways to capture relevant information must be properly justified. The choice of these statistics should be explored in much more details in the manuscript with clear rationale given and caveats with this choice clearly stated.</p><p>2. Related to point 1 – The phylogenetic reconstruction is an unusual way to summarise the statistical data and has not been rigorously assessed. The authors should elaborate on the consequences of the possibly low accuracy of the tree reconstruction itself.</p><p>3. The manuscript is difficult to follow. For example, there is a lengthy part on validating simulations that doesn't have a very specific message and could be reduced quite a lot. Please also ensure that the manuscript narrative is accessible to the general readership of <italic>eLife</italic>, including defining specialist concepts, referring to figures (and figure supplements) in the order they appear in the text. Please also consider adding additional labels to the figures themselves to aid the reader and avoid over-simplification e.g. "Tet" "Dip" could be written in full, Figure 2D is missing a key etc.</p><p>4. More details are required on the ABC analysis, please address the additional comments of reviewer #3 below.</p><p><italic>Reviewer #3 (Recommendations for the authors):</italic></p><p>I focused my attention at looking at the revised ABC analysis.</p><p>I found the new analysis a little hard to follow.</p><p>It is not clear to me precisely what the "step windows" and sliding/expanding timesteps are in Fig6-S1&2. I couldn't see this explicitly explained in the methods. (I can see that the authors are trying to optimise the simulated number of cell divisions, I presume some kind of adaptive search was done)</p><p>I couldn't understand the rationale for only considering the mutation rate in these optimisations. I think that because stronger selection suppresses diversity, the strength of selection should inversely correlate with the time needed to evolve the diversity observed in the organoids.</p><p>I'm concerned about the posterior distributions shown in Figure 6-S3. In most cases the mass in the posteriors is at the extremes of the prior, which suggests that the true parameter values lie outside of the range of the prior (i.e. that the priors were too narrow). This problem appears always the case for the selection coefficient and appears to be a problem for the mutation rate in some cases too (at least it seems that the mutation rate cannot be distinguished from the boundary of 0). Possibly the same issue of a to narrow a prior applies to Figure 5F (taxol treated cell lines). The concern is that the data could be much better explained by a set of parameters located outside of the parameter space considered, which would mean that the conclusions could be incorrect.</p><p>Calculating posterior predictive distributions (analagous to Figure 6-S4&5) would help to convince the goodness of fit.</p><p>Line 479 says that a large-ish value of the acceptance threshold (epsilon=0.05) was used to retain prior data. This seems an incorrect statement: epsilon determines how many simulations are rejected but all prior data is always used. Smaller epsilons should give greater accuracy but at the cost of more computation.</p></body></sub-article><sub-article article-type="reply" id="sa2"><front-stub><article-id pub-id-type="doi">10.7554/eLife.69799.sa2</article-id><title-group><article-title>Author response</article-title></title-group></front-stub><body><disp-quote content-type="editor-comment"><p>Essential revisions:</p><p>1. Modelling.</p><p>Three different models to compute cell fitness based on karyotypic alteration are explored. The construction of all these models feels a little arbitrary, and the assumptions and evolutionary dynamics in each scenario should be more comprehensively explored. Specifically:</p><p>a) In the TOE model, fitness is inversely proportional to average ploidy, so it seems higher ploidies are always selected against. Is this a reasonable assumption? Why is it necessary to divide by the average ploidy?</p></disp-quote><p>This normalization was done to account for gene balance. This fundamental premise of this model is neutral selection if all chromosomes are balanced without excess oncogenes or reduced tumor suppressors. This balance would occur equally with diploid, triploid, and tetraploid cells. By including this normalization, only imbalances which create excess oncogenes or reduce suppressors relative to the entire genome impart selection. We believe this model is most consistent with the known and specific roles of oncogenes and tumor suppressors in promoting or restraining cell proliferation.</p><disp-quote content-type="editor-comment"><p>b) In all models, is the simulated population always "out of equilibrium". If the simulations ran for longer would an "optimal karyotype" be established. Relatedly, the dynamics appear to be strongly influenced by the copy number >6 being lethal – chromosomes (in the TOE model) which are beneficial to be gained might tend to increase copy number to 5 whereas deleterious gains reduce copy number to 1 and the population rests on that "precipice". How reasonable are these "boundary conditions" and do the dynamics change significantly if they are relaxed?</p></disp-quote><p>We believe these boundaries are reasonable as large scale copy number alterations higher than this are rare (see PMID: 31036964 and 32054838). However, to address this, we implemented a variant of our model that considers alternative thresholds. Additionally, we agree that the CIN ON-OFF model had limited biologic relevance and removed this. To improve on this, we have changed our approach to use constant CIN for a much longer period of time (3000 time steps). We agree that WGD is a relevant phenomenon. However, others have already explicitly modeled this (see PMID: 26212324 and 32139907), so we avoid doing the same. Instead, we show that tetraploid founding cells tolerate high mis-segregation rates better than diploid founding cells.</p><disp-quote content-type="editor-comment"><p>c) Gains and losses appear to be treated equivalently – again is this reasonable? Especially in the TOE model where TSG gains and OG losses (and vice versa) have differing consequences. (see also point 8 below).</p></disp-quote><p>While we agree that monosomies are more detrimental than trisomies in non-cancerous tissue, this is not necessarily the case in tumors in which monosomy is often observed (see PMID: 32054838). Nevertheless, to address this critique we have now added a model variant with an additional condition in which cells experience extreme fitness penalties (90% reduction) if any chromosome is haploid. We apply this condition to all selection models and find this attenuates a ploidy increase over time in diploid cells in most selection models (see Figure 3 ‘haploid penalty’).</p><disp-quote content-type="editor-comment"><p>The modelling assumes exponential growth to a relatively small number of cells (4500) and then randomly kills half the cells to reinitiate exponential growth from 2500 cells. This regime will influence the evolutionary dynamics of the system: the random killing will cause emerging clones to often go extinct and could exacerbate the influence of drift in the system. It also effects the influence of selection, (see for example: https://www.nature.com/articles/ng.3214). Alternative growth dynamics could be implemented – such as a Wright Fisher type model of either constant or growing populations (for the construction of a growing WF see: https://pubmed.ncbi.nlm.nih.gov/25665006/) and the influence of the growth dynamics on karyotypic heterogeneity robustly assessed.</p></disp-quote><p>To address these concerns, we improved and more clearly detailed the prior distributions for each inference within the figure legends, we tested for karyotype convergence in each model (see Figure 3), and we demonstrate that inference under the Abundance model is robust to changes in the number of time steps included in the prior data (see Figure 6 — figure supplement 1).</p><disp-quote content-type="editor-comment"><p>The modelling assumes only whole chromosome missegregations (and the bioinformatics data analysis averages over sub-chromsomal sized events). The authors could consider extending their analysis to handle part-chromosome events to better represent the biological data.</p></disp-quote><p>Thank you for this critique. We have now extended the model to handle arm-level segmental aneuploidy.</p><disp-quote content-type="editor-comment"><p>2. Bayesian inference.</p><p>It is a concern that unusual prior distributions are used in the ABC inference and this effects the reliability of the inference. Figure 5F and 6C show smoothed density plots for the prior distributions – which confusingly show density for S<0 – and the true priors might instead be a series of point masses at a handful of S values. This should be clarified.</p></disp-quote><p>We now provide data simulated using uniform distributions of mis-segregation rate and selection.</p><disp-quote content-type="editor-comment"><p>It is likely that the posterior the alteration rate and S are interrelated (high S inferred when the alteration rate is high and vice versa) – so joint posteriors should be shown. Because of the interrelationship between the parameters, it is a concern that the current parameter estimates are inaccurate – currently the prior for S has zero mass for many S values, and the inferred value of the alteration rate will depend on which values of S are explored form the prior distribution. The inference should be repeated using continuous distribution over S, a uniform distribution is suggested.</p></disp-quote><p>We now provide the joint posterior distributions inferred from uniform prior distributions in Figure 6C.</p><disp-quote content-type="editor-comment"><p>When real data is analysed, only the hybrid model is compared to data, but their Figure 3 shows the diversity depends on the underlying model of selection. The authors should implement a model selection routine (one is available with the ABC-sysbio package: https://pubmed.ncbi.nlm.nih.gov/20591907/) to test which selection paradigm (if any) best represents the data. They could also consider comparing what is arguably the "null hypothesis" of neutral evolution (S=0) to a case with selection as part of this model selection, to quantitatively determine the evidence of selection in the data.</p></disp-quote><p>We now include neutral evolution. To address these concerns, we improved and more clearly detailed the prior distributions for each inference within the figure legends, we tested for karyotype convergence in each model (see Figure 3), and we demonstrate that inference under the Abundance model is robust to changes in the number of time steps included in the prior data (see Figure 6 — figure supplement 1).</p><disp-quote content-type="editor-comment"><p>A single threshold for acceptance in the ABC algorithm (epsilon=0.05). The authors should show this is sufficiently small: a straightforward way is to plot the mean and variance of the posteriors as a function of epsilon.</p></disp-quote><p>This is now addressed in the new Figure 6 — figure supplement 1.</p><disp-quote content-type="editor-comment"><p>Above, the possibility of temporal effects in the model are mentioned – how do temporal ("out of equilibrium" evolutionary dynamics) affect the inference?</p></disp-quote><p>To address these concerns, we improved and more clearly detailed the prior distributions for each inference within the figure legends, we tested for karyotype convergence in each model (see Figure 3), and we demonstrate that inference under the Abundance model is robust to changes in the number of time steps included in the prior data (see Figure 6 — figure supplement 1).</p><disp-quote content-type="editor-comment"><p>When public data are analysed, the growth dynamics of the breast tissue, or colorectal organoid is very unlikely to match the exponential growth assumption that is assumed by the authors. These growth dynamics likely strongly determine the pattern of observed heterogeneity (exponential growth leads to large founder effects in the extant population) and so the influence of alterative growth models needs to be explored. This can also be done within a model selection framework.</p></disp-quote><p>We have now concurrently modeled chromosomal instability with a constant population size by approximating constant-population Wright Fisher dynamics (see Materials and methods). We find these models produce similar results at the karyotype level, addressing concerns about the effects of growth patterns on karyotype evolution in this model.</p><disp-quote content-type="editor-comment"><p>The initial state of the model used to fit to public dataset is not specified and needs to be. The authors might also consider the influence of spatial sampling confounders (the breast dataset in particular in not a well-mixed population).</p></disp-quote><p>We have updated the figure legends with more detailed prior and parameter settings.</p><disp-quote content-type="editor-comment"><p>3. Single cell data analysis. The data are presented as if only whole chromosomal alterations occur (e.g. figure 5B). Is this actually the case? (certainly the assumption is false in breast and colon organoid datasets) Could relative read counts in say 1MB bins for each cell be provided in the appendix to reassure the reader that the types of genetic alterations occurring in their single cell data mirror the assumptions of the modelling.</p></disp-quote><p>Paclitaxel causes whole-chromosome aneuploidy through chromosome mis-segregation (Scribano et al. Sci Trans Med 2021). As requested, the binned read counts are now illustrated in the new Figure 5 — figure supplement 2.</p><disp-quote content-type="editor-comment"><p>As noted above, the authors should consider modelling part-chromosome alterations. They should be able to determine in their simulations how their assumption of only whole chromosome missegregation events, despite their being part-chromosomal copy number alterations in the data, affects the accuracy of the inferred chromosomal missegregation rate.</p></disp-quote><p>As requested, we have now performed simulations that allow for segmental aneuploidy.</p><disp-quote content-type="editor-comment"><p>4. The presented framework is lacking expanded characterization and validation of selection models that are biologically relevant. The current framework simply applies a scalar exponent to already published fitness models for selection. It is unclear what this exponent mirrors biologically, beyond amplifying the selection pressures already explored in existing gene abundance and driver density models.</p></disp-quote><p>Biologically, this mirrors the extent to which aneuploid karyotypes are selected for or against, given a particular model. It is an exponent rather than a multiplier because the selection value is already transformed to a probability of division. We now provide model selection and further validation of this method.</p><p>To address this, we have greatly expanded the models and their characterization. We now explicitly include a neutral model throughout, tested various modifications of the model (Figure 3C-E), and use ABC to enable model selection (see Table 3).</p><p>We implemented cellular fitness as the sum of normalized chromosome scores such that the fitness of euploid cells is 1 and the probability of division = 0.5. In this framework, within the ‘abundance’ model, a cell with triploidy of chromosome arm 1p would have a fitness of 0.98. With no additional selection, the probability that this cell divides is 0.98 x 0.5 = 0.49.</p><p>The published fitness models for karyotype selection do not experimentally determine how fitness relates to the probability of division within a given time. For example, there is no clear reason why (or evidence indicating) an extra copy of chromosome arm 1p would reduce the probability of division from 0.5 to precisely 0.49 for a given period. The proposed model of karyotype selection that our ‘abundance’ model is based on only stipulates that aneuploidy of larger chromosomes is more detrimental than small chromosomes. Thus, these fitness values behave as arbitrary units and, therefore, we believe that adjusting and fitting an arbitrary scaling factor to the biological data is appropriate. For example, with an additional selection of S=10, the same cell with trisomy of chromosome arm 1p would divide with a probability of F<sup>S</sup> x 0.5 = 0.98<sup>10</sup> x 0.5 = 0.41.</p><p>We could have implemented a multiplicative framework where fitness (F<sub>mult</sub>) is defined as the total deviation from euploid fitness (1) multiplied by a scaling factor S (F<sub>mult</sub> = S(1 - F)). For the trisomy 1p example, the same fitness value (F<sup>S</sup>=0.98<sup>10</sup>) can be achieved multiplicatively as exponentially via 1 – (9.14 x (1 - 0.98)) ~ 0.98<sup>10</sup>. Thus, the same fitness values can be achieved through arbitrary scaling. We regret that this may have been misinterpreted because it was implemented exponentially vs multiplicatively.</p><p>To further address this critique, we have now better fitted the S values with a flat prior probability across all values, shown how it relates to P<sub>misseg</sub> in posterior probabilities (e.gs, Figure 6C, Table 3) and performed the separate analysis requested.</p><p>The selection values of F are in arbitrary units and so we believe a selection scaling factor is important to include in the model. For example, without additional selection, a hypothetical aneuploid cell with a trisomy resulting in F = 0.95 would be 5% less likely to divide than a euploid cell with F = 1. The exponent scales the selection such that when S = 2, the fitness of the trisomic cell is F ~ 0.9, or 10% less likely to divide. This scaling is necessary to enable both positive and negative selection in a system fitness is decided as the sum of chromosome scores. To further validate the additional weight on selection pressure we did the following:</p><p>1) We constrained the prior distribution of simulated data for our model selection to S=1 giving only the base fitness values without additional scaling. We, again, performed model selection on the data from Bolhaqueiro et al., 2019 and Navin et al., 2011 and found that, with this constrained prior dataset, we inferred mis-segregation rates (see Table 4) that were far below rates seen in cancer cell lines (see Figure 6E).</p><p>2) Given the initial clarification that reviewers were looking for longitudinal analysis, we leveraged data provided by the authors of Bolhaqueiro et al., 2019 where they sequenced single cells from 3 clones from organoid line 16T at 3 weeks and 21 weeks after seeding. We inferred mis-segregation rates and selective pressures in these clones at the 3-week timepoint. We did so under the Abundance model using the same prior distribution of steps given that the diversity of populations under the Abundance model rapidly reach a steady state. When we simulated additional populations using these inferred characteristics we found that the karyotype composition of the simulated populations most closely resembled the biological population than did populations simulated with the unmodified selection values (see Figure 6 — figure supplement 4). This lends credence to the biological relevance of scaled selective pressure vs. unmodified selective pressure.</p><p>We have now concurrently modeled chromosomal instability with a constant population size by approximating constant-population Wright Fisher dynamics (see Materials and Methods). We find these models produce similar results at the karyotype level, addressing concerns about the effects of growth patterns on karyotype evolution in this model.</p><disp-quote content-type="editor-comment"><p>5. Related to point (4)., how is the CIN ON-OFF model in which CIN is turned off after so many cell divisions relevant biologically? Typically, CIN is a considered a trait that evolves later in cancer progression, that once tolerated, is ongoing and facilitates development of metastasis and drug resistance. A more relevant model to explore would be that of the effect of a whole genome duplication (WGD) event on population evolution, which is thought to facilitate tolerance of ensuing missegregation events (because reduce risk of nullisomy).</p></disp-quote><p>We agree that this may not be relevant biologically and have removed the CIN ON-OFF scheme and updated Figure 3 to, instead, explore the convergence of karyotypes. Additionally, we agree that the CIN ON-OFF model had limited biologic relevance and removed this. To improve on this, we have changed our approach to use constant CIN for a much longer period of time (3000 time steps). We agree that WGD is a relevant phenomenon. However, others have already explicitly modeled this (see PMID: 26212324 and 32139907), so we avoid doing the same. Instead, we show that tetraploid founding cells tolerate high mis-segregation rates better than diploid founding cells.</p><disp-quote content-type="editor-comment"><p>5. The authors utilize two models of karyotype fitness – a gene abundance model and driver density model – to evaluate impact of specific karyotypes on cellular fitness. They also include a hybrid model whose fitness effects are simply the average of these two models, which adds little value as only a weighted average. In silico results shows inferred missegregation rates are extremely disparate across the two primary models. And while a description of these differences is provided, the presented analyses do not make clear the most important question – which of these models is more clinically relevant? Toward this, in Figure 2F, the authors claim the three models approach a triploid state – which is unsupported by the in silico results. Clearly the driver model approaches a triploid state, as previously reported. But the abundance model does not and hybrid only slightly so, given that it is simply a weighted average of these two approaches. Because the authors have developed a Bayesian strategy for inferring which model parameters best fit observed data, it would be very useful to see which model best recapitulates karyotypes observed in cancer cell lines or patient materials.</p></disp-quote><p>We agree that the CIN ON-OFF model had limited biologic relevance and removed this. To improve on this, we have changed our approach to use constant CIN for a much longer period of time (3000 time steps). We agree that WGD is a relevant phenomenon. However, others have already explicitly modeled this (see PMID: 26212324 and 32139907), so we avoid doing the same. Instead, we show that tetraploid founding cells tolerate high mis-segregation rates better than diploid founding cells.</p><disp-quote content-type="editor-comment"><p>6. Topological features of phylogenetic trees, while discriminatory, are largely dependent on accurate phylogenetic tree reconstruction. The latter requires more careful consideration of cell linkages beyond computing pairwise Euclidean distances and performing complete-linkage clustering. For example, a WGD event, would appear very far from its nearest cell ancestor in Euclidean space.</p></disp-quote><p>We agree that the abundance and hybrid models are unable to approach a triploid state, in earnest, as does the driver and have made that clearer in the text and improved the figure panel in question for clarity. To address your latter point on which model best fits observed data, we have implemented a model selection scheme to do this (see Table 3). This indicates the gene abundance model as the most biologically relevant and provides evidence for stabilizing selection as the primary mode of selection occurring in the organoid and biopsy data we analyzed.</p><disp-quote content-type="editor-comment"><p>7. Experimental validation of the added selection exponential factor is imperative. Works have already shown models of karyotypic evolution without additional selection exponential coefficient can accurately recover rates of missegregation observed in human cell lines and cancers by fluorescent microscopy. Incorporation of this additional weight on selection pressure has not been demonstrated or validated experimentally. This would require experimental sampling of karyotypes longitudinally and is a critical piece of this manuscript's novelty.</p></disp-quote><p>The selection values of F are in arbitrary units and so we believe a selection scaling factor is important to include in the model. For example, without additional selection, a hypothetical aneuploid cell with a trisomy resulting in F = 0.95 would be 5% less likely to divide than a euploid cell with F = 1. The exponent scales the selection such that when S = 2, the fitness of the trisomic cell is F ~ 0.9, or 10% less likely to divide. This scaling is necessary to enable both positive and negative selection in a system fitness is decided as the sum of chromosome scores. To further validate the additional weight on selection pressure we did the following:</p><p>1) We constrained the prior distribution of simulated data for our model selection to S=1 giving only the base fitness values without additional scaling. We, again, performed model selection on the data from Bolhaqueiro et al., 2019 and Navin et al., 2011 and found that, with this constrained prior dataset, we inferred mis-segregation rates (see Table 4) that were far below rates seen in cancer cell lines (see Figure 6E).</p><p>2) Given the initial clarification that reviewers were looking for longitudinal analysis, we leveraged data provided by the authors of Bolhaqueiro et al., 2019 where they sequenced single cells from 3 clones from organoid line 16T at 3 weeks and 21 weeks after seeding. We inferred mis-segregation rates and selective pressures in these clones at the 3-week timepoint. We did so under the Abundance model using the same prior distribution of steps given that the diversity of populations under the Abundance model rapidly reach a steady state. When we simulated additional populations using these inferred characteristics we found that the karyotype composition of the simulated populations most closely resembled the biological population than did populations simulated with the unmodified selection values (see Figure 6 — figure supplement 4). This lends credence to the biological relevance of scaled selective pressure vs. unmodified selective pressure.</p><disp-quote content-type="editor-comment"><p>8. It seems like this model treats chromosome gains and losses equivalently. Is this appropriate? Chromosome loss events are much more toxic than chromosome gain events – as evidenced by the fact that haploinsufficiency is widespread, and all autosomal monosomies are embryonically-lethal while many trisomies are compatible with birth and development. Can the authors consider a model in which losses exert a more significant fitness penalty that chromosome gains? (see also point 1c above).</p></disp-quote><p>While we agree that monosomies are more detrimental than trisomies in non-cancerous tissue, this is not necessarily the case in tumors in which monosomy is often observed (see PMID: 32054838). Nevertheless, to address this critique we have now added a model variant with an additional condition in which cells experience extreme fitness penalties (90% reduction) if any chromosome is haploid. We apply this condition to all selection models and find this attenuates a ploidy increase over time in diploid cells in most selection models (see Figure 3 ‘haploid penalty’).</p><disp-quote content-type="editor-comment"><p>9. Chromosomes do not missegregate at the same rate (PMID: 29898405). This point should be discussed, and, if feasible, incorporated into the authors' models.</p></disp-quote><p>While this may be true in some contexts, the limited data on this topic (namely Worral et al. Cell Rep. 2018 and Dumont et al. EMBO J. 2020) do not agree on which chromosomes are mis-segregated more often. Worral suggested chromosomes 1-2 are particularly mis-segregated, whereas Dumont finds chromosome 3, 6, X are the highest. These differences may be explained by a context-dependent effects that depend on the model and mechanism of mis-segregation. Worral uses nocodazole washout to generate merotelics whereas Dumont gets mis-segregation through depleting CENP-A. It is unknown which if these mechanisms, if either, is representative of the mechanisms at play in human tumors so we decided to take a general approach assuming equivalent mis-segregation rates. However, we appreciate that this will be a question for other readers and we have now added this to the discussion.</p><disp-quote content-type="editor-comment"><p>10. Can the authors clarify their use of live cell imaging (e.g., in Figure 6G)? Certain apparent errors that are visible by live-cell imaging (like a lagging chromosome) can be resolved correctly and result in proper segregation. Is it appropriate to directly interfer missegregation rates as is done in this paper?</p></disp-quote><p>We did not perform this live cell imaging experiment. We cite these data as being kindly offered by the Kops laboratory and they correspond to the scDNAseq data for normal colon and CRC organoids from Bolhaqueiro et al. Nat Gen. 2019. We agree that chromosome mis-segregation rates cannot be directly inferred by imaging. As you say, lagging chromosomes may resolve and segregate to the correct daughter cell. The fundamental assumption is that, although not all lagging chromosomes mis-segregate, that specimens with higher rate of lagging chromosomes have higher rates of mis-segreation. Because there is no gold-standard measure of CIN in the literature to date, we feel it is necessary to show the correlation between the two and how the data from that study relates to the inferred rates in this study. We have made this clearer in the text.</p><disp-quote content-type="editor-comment"><p>11. The authors should discuss in greater detail earlier mathematical models of CIN, including PMID: 26212324, 30204765, and 12446840. How does their approach improve on this prior work?</p></disp-quote><p>We now provide a more detailed discussion on prior mathematical models, incorporating these and others.</p><p>[Editors' note: further revisions were suggested prior to acceptance, as described below.]</p><disp-quote content-type="editor-comment"><p>The manuscript has been improved but there are some remaining issues that need to be addressed, as outlined below:</p><p>1. The reviewers are concerned about the choice of statistics used for the ABC analysis. While these summary statistics indeed must contain some statistical signal on the rate of chromosomal instability, the current method of "phylogenetic reconstruction" which based on euclidian distances of chromosomal copy number variation is very indirect. The reviewers are not convinced that this is an optimal way to extract relevant information from the distribution of chromosomal copy number across cells. At least they are not clearly rooted in models of clonal evolution.</p></disp-quote><p>We agree that Euclidean distance reconstruction of CNVs is not necessarily the optimal method of phylogenetic reconstruction. The ideal approach would be to use maternal/paternal SNPs to reconstruct phylogeny. However, this is not possible with the data available from state-of-the art single-cell whole-genome sequencing (scWGS). The current state of the art for scWGS is 0.1X coverage so median sequencing coverage is only 10% of the genome. While this is amply sufficient to estimate CNVs across large genomic regions, it is not sufficient to reconstruct phylogenies by SNPs with currently available algorithms. The current work nevertheless makes a material advance over existing studies which use estimates of euclidean distances alone to infer CIN. We have expanded our discussion of this important limitation in the manuscript to acknowledge this limitation.</p><disp-quote content-type="editor-comment"><p>Why this particular set of summary statistics was chosen, how it relates to the quantity of interest (rate of chromosomal instability), how it compares to other possible ways to capture relevant information must be properly justified. The choice of these statistics should be explored in much more details in the manuscript with clear rationale given and caveats with this choice clearly stated.</p></disp-quote><p>Thank you for raising the question about selection of summary statistics. In Approximate Bayesian Computation (ABC), only a small number of summary statistics are employed—larger numbers impair the model (Beaumont et al. Genetics 162: 2025, 2002). To identify the optimal ones, Figure 5 Supplementary 3 evaluates the 9 possible combinations of summary statistics to infer mis-segregation rate, compared with experimental, consistent with what is thought to be the best approach (Csilléry, Katalin, et al. "Approximate Bayesian computation (ABC) in practice." Trends in ecology & evolution 25.7 (2010): 410-418.). Over 90% accuracy is achieved with 4 summary statistics—aneuploidy, MKV, Colless, and cherries—justifying the use of these statistics. This was not well described in the prior version of the manuscript—we have revised the manuscript to clearly describe the approach to select the summary statistics.</p><disp-quote content-type="editor-comment"><p>2. Related to point 1 – The phylogenetic reconstruction is an unusual way to summarise the statistical data and has not been rigorously assessed. The authors should elaborate on the consequences of the possibly low accuracy of the tree reconstruction itself.</p></disp-quote><p>As noted above, we have expanded the discussion about using CNVs to reconstruct phylogenetic trees, noting the possible low accuracy of this tree reconstruction.</p><disp-quote content-type="editor-comment"><p>3. The manuscript is difficult to follow. For example, there is a lengthy part on validating simulations that doesn't have a very specific message and could be reduced quite a lot. Please also ensure that the manuscript narrative is accessible to the general readership of eLife, including defining specialist concepts, referring to figures (and figure supplements) in the order they appear in the text. Please also consider adding additional labels to the figures themselves to aid the reader and avoid over-simplification e.g. "Tet" "Dip" could be written in full, Figure 2D is missing a key etc.</p></disp-quote><p>Thank you for this important critique. As you can see, we have substantively revised the manuscript to improve flow and remove or subsume needless detail in the text. We now ensure that figures and their supplements are in order and have improved labels.</p><disp-quote content-type="editor-comment"><p>Reviewer #3 (Recommendations for the authors):</p><p>I focused my attention at looking at the revised ABC analysis.</p><p>I found the new analysis a little hard to follow.</p><p>It is not clear to me precisely what the "step windows" and sliding/expanding timesteps are in Fig6-S1&2. I couldn't see this explicitly explained in the methods. (I can see that the authors are trying to optimise the simulated number of cell divisions, I presume some kind of adaptive search was done)</p></disp-quote><p>In response to reviews, we had attempted to demonstrate that the model was robust to the number of time steps. Sliding step analysis showed that whether steps 0-10, 10-20, 20-30 etc. were included in the prior, results were similar. Expanding step analysis showed that larger numbers of steps could be included without major impact on the results. Unfortunately, this was a complex analysis and we failed to present it clearly. We have now removed the expanding step windows for simplicity and attempted to simplify the text and display/explain this analysis more clearly. We also have added a relevant section describing this analysis in the methods.</p><disp-quote content-type="editor-comment"><p>I couldn't understand the rationale for only considering the mutation rate in these optimisations. I think that because stronger selection suppresses diversity, the strength of selection should inversely correlate with the time needed to evolve the diversity observed in the organoids.</p></disp-quote><p>Thank you—we have now added selective pressure to these optimizations and display this in Figure 6—figure supplement 1B (to replace the expanding step windows).</p><disp-quote content-type="editor-comment"><p>I'm concerned about the posterior distributions shown in Figure 6-S3. In most cases the mass in the posteriors is at the extremes of the prior, which suggests that the true parameter values lie outside of the range of the prior (i.e. that the priors were too narrow). This problem appears always the case for the selection coefficient and appears to be a problem for the mutation rate in some cases too (at least it seems that the mutation rate cannot be distinguished from the boundary of 0). Possibly the same issue of a to narrow a prior applies to Figure 5F (taxol treated cell lines). The concern is that the data could be much better explained by a set of parameters located outside of the parameter space considered, which would mean that the conclusions could be incorrect.</p></disp-quote><p>We agree that the mass of posteriors should not be close to the edge of the prior. We have revised Figure 5F (now Figure 5E) by expanding the priors to higher mis-segregation rates to address this concern. The updated prior distribution is now used in the revised article. In Figure 6C, the joint inference distributions, S increases asymptotically with low mis-segregation rate. The reason this occurs is because even with infinite selective pressure, at a given rate of chromosome mis-segregation, there will remain a low number of aneuploids in the population. Although this model provides a precise estimate of mis-segregation rate (i.e. CIN), the estimates of selective pressure therefore fall in a very wide band. We infer this to mean that most aneuploid clones are adversely selected against, consistent with prior work (Lukow et al. Dev Cell 2021). We have expanded the description of this in the text.</p><disp-quote content-type="editor-comment"><p>Calculating posterior predictive distributions (analagous to Figure 6-S4&5) would help to convince the goodness of fit.</p></disp-quote><p>Done—see Figure 5—figure supplement 4B.</p><disp-quote content-type="editor-comment"><p>Line 479 says that a large-ish value of the acceptance threshold (epsilon=0.05) was used to retain prior data. This seems an incorrect statement: epsilon determines how many simulations are rejected but all prior data is always used. Smaller epsilons should give greater accuracy but at the cost of more computation.</p></disp-quote><p>Thank you—we have revised this text.</p></body></sub-article></article>