Description:jModelTest2 is a tool to carry out statistical selection of best-fit models of nucleotide substitution
Abstract:This application is jointly developed with Prof. D. Posada (U. Vigo, code owner) and Dr. D. Darriba (U. A Coruña)
The estimation of the evolutionary relationships between DNA sequences obtained from different individuals or species has important biomedical implications. Nowadays, phylogenetic trees are use to predict gene function, to track tumour mutations or to monitor the geographical spread of pathogens, for example. An essential step of the phylogenetic analysis is the selection of appropriate models of nucleotide substitution. During the last years, different statistical selection strategies have been proposed and several programs have been developed to carry out this task; among them, the most popular has been Modeltest [D. Posada, K.A. Crandall. ”Modeltest: testing the model of DNA substitution”. Bioinformatics 14, 817-818 (1998)], which has been superseded by its Java implementation, jModelTest [D. Posada. ”jModelTest: Phylogenetic Model Aver aging”. Mol. Biol. Evol. 25, 1253-1256 (2008)]. Later on, an HPC version was released, jModelTest2 [D. Darriba et al. 2012. jModelTest 2: more models, new heuristics and parallel computing. Nature Methods 9(8), 772.]. It can be freely downloaded from https://code.google.com/p/jmodeltest2/ and allows for the definition of restricted sets of up to 88 candidate models. Among other functionalities, jModelTest is able to implement customizable hierarchical and dynamic likelihood ratio tests, provide a rank of models according to the Akaike Information Criterion, to the Bayesian Information Criterion or to a decision theoretic performance-based approach, calculate the relative importance of every model parameter and compute model-averaged estimates of these, including a model-averaged estimate of the tree topology. By this work, a MPI version to be run on local clusters and a DRMAA release to be executed on Grid or local clusters too are available.