The main objective of this proposal is to develop methods for reconstructing the evolution of polygenic traits using contemporary genetic sequence data. Genome-wide association studies (GWAS) are revealing that most human traits?including traits of biomedical interest?are polygenic, meaning that they are influenced by small-effect variants at many genetic loci. Standard population-genetic methods for reconstructing evolutionary history are underpowered for polygenic traits?when there are many variants of small effect, signatures of natural selection are spread across the genome and subtle at any one locus. Dr. Coop (sponsor) and others have developed methods for detecting selection on polygenic traits, but existing methods are limited: they do not reveal the timing and strength of selection, and they do not reveal the course of either selected or neutral change in the trait over time. The applicant will develop methods that improve on existing approaches by working with ancestral recombination graphs (ARGs), a rich representation of a sample?s history of recombination and common-ancestor events across the genome. ARG estimation from whole- genome data is improving rapidly, and ARG-based approaches are positioned to benefit from these advances.
AIM 1 will produce statistical methods that estimate a population?s historical mean values for a trait?s polygenic score, as well as statistical tests to identify trait histories that are inconsistent with neutral evolution. These procedures will be assessed for robustness in extensive simulations.
AIM 2 will apply the methods of Aim 1 to human genomic data using GWAS effect sizes and genomic data from the 1000 Genomes project. The applicant will seek to replicate and refine previous results suggesting recent natural selection on height in Europe and examine many other phenotypes. The applicant will receive training in the population genetics of natural selection, quantitative genetics, and the processing and analysis of whole-genome data. The sponsor and the institutional environment are well-suited to support the applicant in this training. The new skills gained by the applicant will equip him to answer questions about polygenic traits throughout a research career in statistical population genetics.
Statistical geneticists are converging on the conclusion that most human traits, including traits of biomedical interest, are polygenic?that is, influenced by many genetic loci of small effect. Polygenic adaptation proceeds via small, coordinated changes in allele frequency at many genetic loci, rendering it difficult to understand with existing population-genetic methods. In my postdoc, I will develop and apply methods for reconstructing the evolution of polygenic traits, capitalizing on cutting-edge algorithms for representing whole-genome data using ancestral recombination graphs (ARGs).
