Métodos de reconstrucción filogenética II: inferencia bayesiana

Phylogenetic reconstruction through Bayesian inference is currently widely used. The main advantage of this method is the direct output of posterior probabilities for each clade on the final phylogeny. Thus, it does not require bootstrapping as a measure of uncertainty. Moreover, Bayesian inference is perfectly fit for dating phylogenies through molecular clocks. In this paper, the basics of Bayesian inference applied to phylogenetic reconstruction are described, starting with an explanation of Bayes’ theorem. Then, the use of the Metropolis-Hastings algorithm to sample topologies from the posterior distribution is characterized and illustrated through a simple example. At the end, there is a mention of the software used for Bayesian phylogeny reconstruction.