Genomic patterns of divergence and gene flow in Drosophila and Goodeidae

Abstract

Understanding the impact of the presence or absence of geographic barriers on speciation has been a source of enduring interest in evolutionary biology. The degree of geographical isolation and the prevalence of gene flow during divergence may often determine the rate of reproductive isolation and the relative importance of other evolutionary forces. In this thesis, I use comparative genomic approaches to ask how frequent gene flow is, understand the contributions of ancient and recent gene flow to divergence, and finally, examine how patterns of genetic divergence between closely-related species are affected by geography and gene flow in two different systems: Drosophila and Splitfins (Goodeidae). I investigate the evolutionary history of divergence in the virilis group of Drosophila using de novo whole-genome sequences. I show that ancient and recent gene flow is common complicating phylogenetic inference and confounding levels of genetic divergence on the X chromosome across species pairs in the group. Next, I test whether gene flow is common across 96 species pairs in Drosophila. Using model-based inference from whole-genome data, I demonstrate that both allopatric and sympatric species pairs in Drosophila show similar support for models of speciation-with-gene-flow. Additionally, using inferred migration and divergence time estimates for species pairs across Drosophila, I show that evidence for reinforcement is mixed. In the final two chapters, I examined patterns of divergence and gene flow in Goodeidae. Using whole-genome sequences and de novo assemblies, I show that divergence of Goodeidae occurred rapidly, alongside consistent fluctuations in population size and limited ancient gene flow. Finally, I employ a broad comparative genomic approach using 21 genomes across Cyprinodontiformes to identify signatures of molecular convergence and positive selection linked to the evolution of viviparity in Goodeidae and Poecilidae. Altogether, this thesis demonstrates that complex speciation histories with gene flow are the rule rather than the exception.