Bayesian inference of natural selection from time series data of allele frequencies across linked loci

题目:Bayesian inference of natural selection from time series data of allele frequencies across linked loci

报告人:Feng Yu,School of Mathematics,University of Bristol

时间:2018.12.25     9:30-10:30


摘要:The rapid improvement of DNA sequencing technology has made it possible to monitor genomes in great detail over time. This presents an opportunity to investigate natural selection based on time series DNA data while accounting for genetic recombination. Such time series data allow for more precise estimates of population genetic parameters and hypothesis testing on recent actions of natural selection. Here we develop a novel Bayesian statistical framework for the inference of natural selection by capitalising on the temporal aspect of DNA data across linked loci, regardless of whether the genotype data have been phased or not. Our approach relies on a hidden Markov model incorporating the two-locus Wright-Fisher diffusion with selection, which enables us to explicitly model the process of genetic recombination. The posterior probability distribution for the selection coefficients is obtained with the particle marginal Metropolis-Hastings algorithm. The performance of our method is evaluated through extensive simulations. We show that our estimates for the selection coefficients are unbiased. Moreover, under certain circumstances, we find that our approach can deliver precise estimates for the selection coefficients whereas existing single-locus methods fail, especially when the two loci are tightly linked. The utility of our approach is illustrated with an application to linked loci encoding coat colour in horses, where we also show that our method is capable of handling missing values in DNA data.

报告人简介:Dr Feng Yu received a Ph.D. in Mathematics from the University of British Columbia, Canada in 2005, under the supervision of Edwin Perkins. His Ph.D. work focused on probabilistic modelling of ecology and biology. From 2005 to 2007, he was a post-doctoral fellow in the Department of Statistics at the University of Oxford, UK, where he worked on probabilistic modelling of natural selection. Since 2007, he has been lecturer, then senior lecturer, in the School of Mathematics at the University of Bristol, UK. His current research interest mainly lies in statistical genetics, especially modelling and inference of selection.