Phylogenetic Likelihood Equations: What Do They Mean and Why Should We Care?

• Activity Number: 351
• Type: Contributed
• Date/Time: Tuesday, July 31, 2012 : 10:30 AM to 12:20 PM
• Sponsor: WNAR
• Author(s): Vladimir N Minin*+
• Companies: University of Washington
• Keywords: phylogenetics ; estimating equations ; stochastic processes ; incomplete data ; evolution

Abstract

Phylogenetic likelihood is defined as a probability of observing discrete evolutionary traits (DNA, amino acids, etc.) at the tips of a phylognetic tree. The underlying unobserved evolution of the trait of interest is modeled as a continuous-time Markov chain. Although great numerical advances have made over the last thirty something years to efficiently evaluate phylogenetic likelihood, this function is often treated as a black box. Here, we try to look under the hood of the phylogenetic likelihood and to understand what it is that we do when we maximize this function. We show that interesting insights can be obtained by viewing phylogenetic likelihood as the likelihood of a partially observed exponential family model. We demonstrate how this new point of view can be used for deriving alternative estimators and for assessing goodness of fit of phylogenetic models.