An Extended Ancestral Mixture Model for Phylogenetic Inference under the HKY DNA Substitution Model

Abstract

An Extended Ancestral Mixture Model for Phylogenetic Inference under the HKY DNA Substitution Model Shu-Chuan Chen Department of Statistics, Idaho State University, USA ? Inferring phylogenies is important in many scientific areas such as biology, paleontology, biochemistry and bioinformatics. Many statistical methods have been proposed to reconstruct the phylogeny from the DNA sequences, such as Maximum Parsimony method, Distance method, Maximum Likelihood method and Bayesian method. Recently, a new method, Mixture Ancestral Models has been proposed by Chen and Lindsay (2006) to infer phylogeny from the binary DNA sequences. In the talk, we extend the ancestral mixture models to the one which can infer phylogeny from the DNA sequences with multiple states. By varying the time parameter, one can create a hierarchical tree that estimates the population structure at each fixed backward point in time. Theoretical and computational properties and applications of the extended ancestral mixture model will be presented. This is a joint work with Dr. Jay Taylor at Arizona State University.