I am using nucleotide sequences for baseyan phylogenetic analysis. The program MrBayes involves the parameter of specifying the model. Does this tool have any option of automatic selection of best fit model as in other ML programs? If no, how to determine which model to use?
Thanks
2 answers
For nucleotide sequences, jModelTest can be used to determine the best model.
Following criteria are used for selection of the best-fitting models: a) the Akaike Information Criterion (AIC), b) the Bayesian Information Criterion (BIC) and c) hierarchical Likelihood-ratio tests (hLRTs) Tools like 1. MEGA7 2.ModelGenerator (https://bmcevolbiol.biomedcentral.com/articles/10.1186/1471-2148-6-29) 3. ProtTest (https://academic.oup.com/bioinformatics/article/21/9/2104/408994) - only for proteins
Furthermore, Miguel Arenas summarized all softwares that available for selection of best-fitting models (https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4620419/).
According to you requirement you can choose a model and use that model for MrBayes.
Log in to answer this question.