Bayesian phylogenetics

Bayesian phylogenetics estimates probability distributions over evolutionary trees by combining sequence or trait data with prior assumptions and models of evolution.

Bayes' theorem: A rule for updating probabilities by combining prior beliefs with the likelihood of observed evidence. It is the probability-update rule underlying posterior probabilities for trees and model parameters.
Molecular phylogenetics: The study of evolutionary relationships using DNA, RNA, or protein sequences. Bayesian methods are widely used to infer trees from molecular sequence alignments.
Molecular clock: A model relating genetic change to elapsed time, often used to estimate divergence dates. Clock models let Bayesian analyses estimate when lineages diverged alongside their relationships.
Maximum likelihood estimation: A method that selects parameter values maximizing the probability of observed data under a model. It typically reports a best-fitting tree rather than a posterior distribution over trees.
Likelihood function: A function measuring the probability of observed data under specified model parameters. Sequence likelihoods score how well a candidate tree and evolutionary model explain the data.
Sequence alignment: An arrangement of biological sequences that identifies positions treated as evolutionarily corresponding. The alignment defines which character observations contribute to the tree analysis.
Bayesian dating: The estimation of evolutionary divergence times using Bayesian models, often with fossil or geological calibrations. It combines tree inference with calibrated time estimates and uncertainty distributions.
Maximum parsimony: A tree-inference criterion that favors evolutionary histories requiring the fewest character changes. It avoids explicit probability distributions over trees and evolutionary parameters.
Phylogenetic tree: A branching diagram representing hypothesized evolutionary relationships among organisms, genes, or other entities. The tree topology and branch lengths are central unknowns in the Bayesian analysis.
Substitution model: A probabilistic model describing how DNA, RNA, or protein characters change over evolutionary time. The likelihood depends on assumptions about substitutions along the tree's branches.

Linked from 11 pages