Recent Advances in the Inference of Gene Flow from Population Genomic Data

Abstract

Detecting gene flow between populations or species is a fundamental goal of population genetics and speciation research and is also central for a thorough understanding of the demographic history of lineages. While population genomic data offer an unparalleled opportunity to study gene flow and other evolutionary processes at high resolution, extracting meaningful patterns from such large and complex datasets is rarely straightforward. Recent advances in both theory and methodology have led to a number of newly proposed analytical tools and frameworks for inferring genome-wide patterns of introgression and admixture that can more efficiently leverage population genomic data. Here, we provide an overview of several recent contributions to the problem of estimating gene flow, discuss advantages and potential pitfalls to these approaches, and provide an outlook for future developments. Three prominent areas of recent research progress include (1) improving upon existing test statistics to detect and measure gene flow, (2) developing efficient frameworks for demographic model testing, and (3) applying supervised machine learning to identify introgressed loci across genomes. Over the past several years, contributions to these three areas have greatly enhanced our ability to study gene flow at various scales (i.e., species, populations, and individual genomes). Here, we highlight six relevant studies within these focal areas that represent particularly novel contributions to the goal of gene flow estimation from genome-scale data. The inference of gene flow is a notoriously challenging statistical problem that is an integral component of population genomic research. Our survey of the literature revealed a number of important recent contributions to this problem, from the improvement of admixture tests to demographic model testing and inference of specific regions of the genome likely to have crossed boundaries between populations and species. Although these studies represent only a sampling of the current literature, their contributions, along with those from numerous studies in the expanding field of population genomics, are markers of considerable progress in recent years toward addressing the issue of genomic inference of gene flow.