Simultaneous Inference of Substitution Rates Across Sites in DNA Sequences

• Activity Number: 71
• Type: Contributed
• Date/Time: Sunday, August 3, 2014 : 4:00 PM to 5:50 PM
• Sponsor: Biometrics Section
• Author(s): Tingting Zhao*+ and Alexandre Bouchard-Côté
• Companies: University of British Columbia and UBC
• Keywords: Phylogeny ; Maximum Likelihood ; EM Algorithm ; Rate Factor

Abstract

Different regions of the same gene can have different rates of mutations. Many attempts have been tried to allow for variable rates of substitution across nucleotide sites. A rate factor from a continuous distribution has been introduced but can only scale to less than a dozen sequences for computational reasons. The main contribution of our work is that we propose a discrete distribution over the rate factor and a new implicit model which combines the General Time Reversible (GTR) model, the rate factors and the distribution of rate factors into a rate matrix parameterized by weights and features. Within this framework, we also develop a hierarchical model to incorporate the non-local interactions in protein secondary structure into amino acid substitution process. Our model is more flexible while preserving attractive computational properties. We can estimate both the rate matrix and the rate factors simultaneously via an Expectation Maximization algorithm efficiently. We evaluate our method by both simulations and a real dataset. From the real dataset, it reflects that the method is useful for large phylogenies with thousands of sequences.