PhyML

PhyML is a software package for phylogenetic inference using the maximum likelihood (ML) method. It estimates phylogenetic trees from aligned nucleotide or protein sequence data and is widely used in evolutionary biology. PhyML was developed by Philippe Guindon and Olivier Gascuel, with continued updates that expanded model options and search strategies.

The program supports a range of substitution models for nucleotides (including GTR, HKY, and JC69) and for proteins (such as JTT, WAG, and LG). It can incorporate among-site rate variation with a discrete gamma model and, optionally, a proportion of invariant sites. PhyML uses heuristic tree-search strategies, primarily Nearest Neighbor Interchange (NNI) and, in newer versions, Subtree Pruning and Regrafting (SPR), to explore tree space. It can perform automatic model selection through Smart Model Selection (SMS) and report model fit using criteria such as AIC or BIC. The software outputs a maximum likelihood tree with branch lengths, along with estimated parameters and the log-likelihood. It can assess branch support via approximate likelihood ratio tests (aLRT) or bootstrap analyses.

PhyML is available as a standalone command-line program and is also accessible via a web server as part of the Phylogeny.fr platform. It is distributed under a free/open-source license and is implemented in C/C++. It accepts standard alignment formats (FASTA, PHYLIP) and provides options for bootstrapping, multiple runs, and parallel execution on compatible systems.