Computational Statistical Inference for Molecular Evolution and Population Genetics

Abstract

This research aims to develop new methods and software for evolutionary inference. The focus will be on two challenges that analysis of molecular data in the genomic age provides: (1) measurably evolving populations and (ii) evolution of RNA secondary structure. Molecular sequence data is increasing in length, and also acquiring a depth in the rime dimension (for example, HIV-1, human influenza A, and ancient mtDNA). This has provided an innovative research direction, for which explicit evolutionary inference methods are required. The first aim of this research is to provide new statistical methods and new bioinformatic tools (software packages) to assist in tackling this new problem in evolutionary biology. Both maximum likelihood and Bayesian inference methods are developed for the purpose of estimating substitution rates and concerted changes in the substitution rate. In addition, with the rapid succession of newly sequenced full genomes, researchers can no longer use simple molecular sequence similarity to infer homology. Knowledge of molecular structure needs to be incorporated into evolutionary inference methods. The evolutionary relationship between sequence and structure is still poorly understood and the new wealth of data provides an exciting opportunity to guide theoretical developments. The second major objective of this research is to use the wealth of sequence data available to explore the role and impact of RNA secondary structure on evolution. To this end, empirical studies and simulations are undertaken to explore the role of RNA secondary structure in the evolution of 16S-like rRNA-encoding genes. Finally the inference of spatially resolved populations from gene sequences is briefly investigated. This research project has both computational and conceptual objectives. In both cases, the concrete result of these objectives will be new statistical models and computer software for evolutionary inference and a better understanding of the action of molecular and population processes during evolution.