Wed 9 Mar. 2011, 14:00
Title: Bayesian inference of ancient human demography from individual genome sequences
Speaker: Adam Siepel, Cornell University
Complete genome sequences are now available for individuals from several major human population groups. Here we describe an effort to estimate key evolutionary parameters from the complete genome sequences of six individuals from six different populations. Employing a Bayesian approach based on coalescent theory, we extracted information about ancestral population sizes, divergence times, and migration rates from inferred genealogies at many neutrally evolving loci from across the genome. To analyze data from human individuals, we modified previous methods to account for gene flow between populations and to integrate over possible phasings of diploid genotypes. We also developed a custom pipeline for genotype inference to mitigate possible biases from heterogeneous sequencing technologies, coverage levels, and read lengths. We estimate that the most recent common ancestral population of modern humans lived 108-157 thousand years ago (kya) and had an effective size of ~9,000, and that Eurasian and West African populations diverged 38-64 kya.