Improving the Study of Gene Rearrangement in Bayesian Phylogeny

Xiting Yang

Defense Date

7-22-2004

Graduation Date

Summer 1-1-2004

Availability

Campus Only

Submission Type

thesis

Degree Name

MS

Department

Computational Mathematics

School

McAnulty College and Graduate School of Liberal Arts

Committee Chair

Donald L. Simon

Committee Member

Francesco Cesareo

Committee Member

Kathleen A. Taylor

Committee Member

Patrick Juola

Keywords

Bayesian Phylogeny, Markov chain Monte Carlo, Three Neighbors of an Internal Node

Abstract

Examining gene rearrangement is one important method for constructing phylogenies, or evolutionary histories, of taxa. Our research uses Bayesian approach, implemented by a Markov chain Monte Carlo algorithm, for finding the most probable phylogenies and assessing their likelihood given gene arrangement data. In past research, when updating trees, algorithms examined the arrangement data stored in pairs of connected nodes. We develop a new method, which examines the arrangements of the three neighbors of an internal node. The rearrangements found by the method are not guaranteed to be optimal, but will, with high probability, be close to optimal. Initial tests show that the new update method improves the MCMC procedure.

Format

PDF

Language

English

This document is currently not available here.

Share

COinS