New Computational Approaches to Study the Evolution of Asexual Haploids

Abstract

Numerous factors can influence the evolutionary fate of mutations. Despite this, we tend to study strong evolutionary drivers, or evolution under simple contexts, in part because they are the conditions we have a means to study. My thesis evaluates novel computational approaches to advance detection, and study, of factors that influence a mutation’s evolutionary outcome. First, I present the novel computational tool AEGIS that I use to detect phylogenetic signals of correlated evolution followed by an experimental approach to evaluate the role of epistasis as a potential cause of correlated evolution among sites associated with antibiotic resistance in Pseudomonas aeruginosa. Second, I developed rSHAPE, a novel in silico approach for experimental evolution with asexual haploids, to complement empirical work by providing a common framework in which to test various evolutionary scenarios. After demonstrating that rSHAPE replicates the expected evolutionary dynamics of de novo mutations, I provide evidence that the common laboratory practice of serial passaging may increase stochasticity of evolutionary outcome. Through my work, I have demonstrated that a marriage of computational and experimental approaches will offer new opportunities to understand how the interaction of evolutionary factors influence the fate of mutations.