[ˈbeɪzɪən]
Bayesian methods are a general approach to statistical inference where observations are combined with formally specified a priori knowledge, or prior for short. Inference is performed by probability calculus and the result is an updated, a posteriori, state of knowledge, or posterior for short. Both the prior and the posterior are represented as probability distributions over different states of affairs. In phylogenetics, the states of affairs over which the prior and posterior are given typically correspond to different phylogenetic tree structures and the associated parameter values.
The advantages of Bayesian phylogenetic methods include a principled way to express prior knowledge, automatic quantification of uncertainty, and the possibility to combine different data sources that may, e.g., describe different subsets of the taxa, or different data types such as sequence data and phenotypic observations.
A popular software package for Bayesian phylogenetic inference is MrBayes. See tools.
In other languages
DE: Bayessche Phylogenetik
FR: phylogénie bayésienne
IT: filogenesi Bayesiana
TR