Markov chain Monte Carlo and its applications to phylogenetic tree construction

Abstract

This thesis addresses the application of Bayesian methods to problems in phylogenetics. Specifically, we focus on using genetic data to estimate phylogenetic trees representing the evolutionary history of genes and species. Knowledge of this common ancestry has implications for the identification of functions and properties of genes, the effect of mutations and their roles in particular diseases, and other diverse aspects of the biology of cells. Improved algorithms for phylogenetic inference should increase our potential for understanding biological organisms while remaining computationally efficient. To this end, we formulate a novel Bayesian model for phylogenetic tree construction based on recent studies that incorporates known information about the evolutionary history of the species, referred to as the species phylogeny, in a statistically rigorous way. In addition, we develop an inference algorithm for this model based on a Markov chain Monte Carlo method in order to overcome the computational complexity inherent in the problem. Initial results show potential advantages over methods for phylogenetic tree estimation that do not make use of the species phylogeny.