Demographic and Population Separation History Inference Based on Whole Genome Sequences.

population history;whole genome sequence;dog domestication;human evolutionary history;gorilla evolutionary history;Genetics;Science;Bioinformatics

Song, ShiyaKang, Hyun Min ;

University of Michigan

关键词: population history;    whole genome sequence;    dog domestication;    human evolutionary history;    gorilla evolutionary history;    Genetics;    Science;    Bioinformatics;   

Others  :  https://deepblue.lib.umich.edu/bitstream/handle/2027.42/133341/songsy_1.pdf?sequence=1&isAllowed=y

瑞士|英语

来源: The Illinois Digital Environment for Access to Learning and Scholarship

【 摘 要 】

Patterns of DNA sequence variation among present day individuals contain rich information about past population history. The recent availability of whole genome sequences provides challenges and opportunities for developing computational methods to infer detailed models of population history. The goal of this thesis is to extend current methodology and apply available techniques to answer questions about population history in human, gorilla and canine species.Recent methodologies based on the sequentially Markovian coalescent model permit the inference of population history using single or several whole genome sequences. However, these approaches fail to generate parametric estimates for split times, which are confounded by subsequent migration. Additionally, the effect of switch errors resulted from statistical phasing on split time estimation is largely unknown. We reconstructed phased haplotypes of nine individuals from diverse populations using fosmid pool sequencing. We analyzed population size and separation history using the Pairwise Sequentially Markovian Coalescent model (PSMC) and Multiple Sequentially Markovian Coalescent model (MSMC) and found that applying MSMC on statistically phased haplotypes results in more recent split time estimation compared with physically phased haplotypes due to switch errors. We further extended PSMC with Approximate Bayesian Computation to infer split time and migration rates under a standard isolation with migration model. We dated several key events in human separation history using these methods.Gorillas are human’s closet living relatives other than chimpanzees. We analyzed whole genome sequencing data of thirteen gorilla individuals and applied GPhoCS, a Bayesian coalescent-based approach to infer ancestral population sizes, divergence times and migration rates amongst three gorilla subspecies, shedding light on the evolutionary forces that have uniquely influenced patterns of gorilla genetic variation.The origins and dynamics of dog domestication has been a controversial and intriguing problem. We analyzed two ancient dog genomes from the Neolithic and over 100 contemporary canine genomes. While both dogs show signatures of admixture, they predominantly share ancestry with modern European dogs, contradicting a late Neolithic population replacement suggested by mitochondrial studies. By calibrating the mutation rate using our oldest dog, we narrowed the timing of dog domestication to a window of 20-40 kyrs ago.