Buzbas Erkan Ozge, Verdu Paul
Department of Statistical Science, University of Idaho, United States.
CNRS/MNHN/Université Paris Diderot/Sorbonne Paris Cité, France.
Theor Popul Biol. 2018 Jul;122:149-157. doi: 10.1016/j.tpb.2018.03.006. Epub 2018 Mar 28.
Signatures of recent historical admixture are ubiquitous in human populations. We present a mechanistic model of admixture with two source populations, encompassing recurrent admixture periods and study the distribution of admixture fractions for finite but arbitrary genome size. We provide simulation-based methods to estimate the introgression parameters and discuss the implications of reaching stationarity on estimability of parameters when there are recurrent admixture events with different rates.
近期历史混合的特征在人类群体中普遍存在。我们提出了一个包含两个源群体的混合机制模型,涵盖了反复混合的时期,并研究了有限但任意基因组大小下混合比例的分布。我们提供了基于模拟的方法来估计基因渗入参数,并讨论了在存在不同速率的反复混合事件时达到平稳状态对参数可估计性的影响。