Bayesian phylogenetic inference using DNA sequences: a Markov Chain Monte Carlo Method

An improved Bayesian method is presented for estimating phylogenetic treesusing DNA sequence data. The birth-death process with species sampling isused to specify the prior distribution of phylogenies and ancestralspeciation times, and the posterior probabilities of phylogenies are usedto estimate the maximum posterior probability (MAP) tree. Monte Carlointegration is used to integrate over the ancestral speciation times forparticular trees. A Markov Chain Monte Carlo method is used to generate theset of trees with the highest posterior probabilities. Methods aredescribed for an empirical Bayesian analysis, in which estimates of thespeciation and extinction rates are used in calculating the posteriorprobabilities, and a hierarchical Bayesian analysis, in which theseparameters are removed from the model by an additional integration. TheMarkov Chain Monte Carlo method avoids the requirement of our earliermethod for calculating MAP trees to sum over all possible topologies (whichlimited the number of taxa in an analysis to about five). The methods areapplied to analyze DNA sequences for nine species of primates, and the MAPtree, which is identical to a maximum-likelihood estimate of topology, hasa probability of approximately 95%.