Bayesian Hierarchical Graph Models for Phylogeography

• Activity Number: 181
• Type: Contributed
• Date/Time: Monday, July 30, 2012 : 10:30 AM to 12:20 PM
• Sponsor: Section on Bayesian Statistical Science
• Author(s): Gabriela Cybis*+ and Marc A Suchard and Janet Sinsheimer
• Companies: University of California at Los Angeles and University of California at Los Angeles and University of California at Los Angeles
• Keywords: Bayesian Phylogenetics ; Hierarchical Model ; Phylodynamics

Abstract

Viral pathogens kill every year millions of people worldwide. Accurate information on their dispersion process is important for the formulation of better public health policies. Bayesian phylogeographic methods integrate geographic and evolutionary modeling to assess such information. We enhance the potentialities of existing phylogeographic methods by integrating multiple datasets in a hierarchical setting. In our model there are n exchangeable strata consisting of viral sequences and locations; each evolving along a phylogenetic tree and according to a conditionally independent geographic process. At the hierarchical level, a random graph summarizes the overall dispersion process by informing which migration rates between sampling locations are likely to be nonnegative in the strata. This approach can then be used to test differences and commonalities between strata. Our method represents an appropriate framework for analyzing inherently hierarchical datasets, such as multilocus data. We use a data set consisting of multiple subtypes of Dengue Virus in the Americas to showcase the potentialities of our method.