Applications of Coalescent Theory to Introgression Inference and Analysis of Genomic Studies of Sex and Sexuality

Abstract

Coalescent theory uses a backwards-in-time approach to model the genetic ancestry of a sample. The individuals in the sample can be imagined as residing in the same totally mixed population, as in the standard coalescent model, or in different, potentially isolated populations, as in the multispecies coalescent. It is becoming increasingly evident, however, that species are not entirely isolated, with gene flow, or introgression, occurring across species boundaries. Characterizing the effects and signal of this permeability is vital for understanding how species evolve in nature. In chapter one, I develop a statistical technique that uses the internal branch lengths of gene trees to infer what portions of the genome may be introgressed. Using simulations, I show that the developed method infers the presence of introgression even under models of population size change and recombination, and, with collaborators, apply it to look for introgressed regions in data from Heliconius butterflies. In chapter two, I present a theoretical study of the joint distribution of gene trees at linked loci in the presence of introgression. Coupled with simulations, I use this model to show that the outcomes of incomplete lineage sorting at linked loci are informative to the true evolutionary history. In chapter three, I explore a connection between recombination rate and a technique from applied topology called persistent homology. With collaborators, I show that various features of the barcode, an encoding of the shape of the data at different scales, of sequences sampled from a population are correlated with recombination rate. Finally, in chapter four, I present interdisciplinary work with the GenderSci lab that places genetic studies of human sex and sexuality under a critical lens and explores possible sites for intervention and critique by gender and sexuality scholars.