A Nonparametric Wavelet-Based Approach to Modeling the Genetic Diversity of Influenza A Through Time

• Activity Number: 121
• Type: Topic Contributed
• Date/Time: Monday, August 2, 2010 : 8:30 AM to 10:20 AM
• Sponsor: Section on Bayesian Statistical Science
• Author(s): Jennifer Aileen Tom*+ and Marc Suchard and Janet Suzanne Sinsheimer
• Companies: University of California, Los Angeles and University of California, Los Angeles and University of California, Los Angeles
• Keywords: Bayesian ; Phylogenetics ; Wavelets

Abstract

The computational burden of estimating a Bayesian phylogenetic model often requires partitioning the dataset, by location for example, and inferring the phylogeny of each partition independently. The computational burden increases when assuming the "skyline plot," a piecewise-constant demographic model producing samples of genetic diversity through time. The compromise to partition the dataset into more tractable sizes results in samples from intermediate distributions dependent only on the data within a given partition and removed from the context that justified the initial data collection. We demonstrate recycling of these samples into a unified Bayesian nonparametric wavelet based model allowing formal statistical inference based on the entire dataset. We apply this methodology to twelve years of influenza A data, a virus whose evolutionary dynamics are well captured by wavelets