Maximum Likelihood Phylogenetic Inference

Abstract

In this article, we provide an overview of maximum likelihood methods for phylogenetic inference. A brief introduction to general maximum likelihood estimation is provided. We define a phylogenetic likelihood, summarize how to compute this likelihood, and then discuss approaches used to maximize the phylogenetic likelihood function. We discuss a property of the maximum likelihood estimation, called consistency, that states that the maximum likelihood phylogeny will converge to the true phylogenetic tree with as more and more data are added to the analysis. We describe the bootstrap, a popular technique used to characterize the uncertainty in parameter estimates, and then outline its use in phylogenetic maximum likelihood estimation. A short example is given to illustrate the use of phylogenetic maximum likelihood techniques on a real dataset of primate mitochondrial DNA sequences.