When Does Gene Flow Stop? A Mechanistic Approach to the Formation of Phylogeographic Breaks in Nature

Abstract

I present a dissertation that examines the stochastic and deterministic predictors of population genetic demes between populations of organisms in nature. Gene flow patterns can increase the demographic viability of a population by increasing genetic diversity, and therefore reducing inbreeding depression and improving the population’s ability to adapt to changes in their biotic or abiotic environment. Conversely, gene flow patterns can negatively impact demographic health by flooding locally adapted phenotypes and reducing the population’s overall fitness, causing outbreeding depression. On the species level, disruptions in gene flow allow species to accumulate a variety of adaptations, and therefore provides the underlying variation necessary for the first step in diversification. In my dissertation, I use a variety of empirical and theoretical approaches to examine the mechanisms behind reductions in gene flow between natural populations. I propose that there are three common contributors to a breakdown in gene flow between populations: 1) the existence of an abiotic barrier that reduces migration by imposing intractable physiological costs on organisms that try to cross it, 2) ecological or behavioral properties of a population that reduce dispersal, such as reluctance to cross open areas that the species is still physically able to move through or 3) failure of dispersing individuals to survive in a new environment long enough to breed, possibly due to interspecific interactions, including predation and parasitism. The mechanisms behind these barriers are not mutually exclusive. Further, signatures of low gene flow between natural populations can reflect both deterministic barriers and stochastic patterns of movement of alleles across a landscape. Untangling all of this complexity requires a variety of theoretical and empirical approaches. I present empirical methods to identify geographic barriers that reduce gene flow between populations, and to identify cases in which genetic deme boundaries are not related to abiotic or biotic barriers but instead reflect stochastic patterns of migration. For this work I use night lizards from the genus Xantusia, including Xantusia vigilis and Xantusia riversiana. I use the discordance between SNP data derived from a ddRAD sequencing and microsatellite markers to better understand the demographic circumstances that result in the maintenance of phylogeographic breaks in the absence of a clear barrier to dispersal. I use an individual-based stochastic simulation inspired by natural polymorphic systems including Sonora snakes, Oophaga dart frogs, and Heliconius butterflies to identify mechanisms behind reduction gene flow due to predation. I propose a pipeline to use existing next-generation sequence data to integrate parasite biogeographic patterns into examinations of the mechanisms behind host population genetic patterns. My work on the complex and contingent nature of barriers to gene flow suggests that a hypothesis-testing framework, in which a suite of potential mechanisms are sequentially ruled out by the available evidence, might be the most productive approach to understanding patterns of genetic diversity in nature. Such an approach would be useful in conservation genetics, particularly in planning to maintain corridors for dispersal. Further, I suggest that improved understanding of the physiological tolerance of species of interest, and improved understanding of their biological context, can improve our predictions about the types of scenarios that lead to reductions in gene flow and ultimately set populations on the path to diversification.