Maximum clade credibility tree

A maximum clade credibility tree is a tree that summarises the results of a Bayesian phylogenetic inference. Whereas a majority-rule tree combines the most common clades, potentially resulting in a tree that was not sampled during the analysis, the maximum-credibility method evaluates each of the sampled posterior trees. Each clade within the tree is given a score based on the number of times that it appears in other sampled posterior trees, and these scores are added to give a total score for the tree. The tree with the highest score represents the maximum clade credibility tree.^[1]

Since each clade's score is akin to a probability, it may be more appropriate to multiply, rather than add, each clade's score (expressed as a probability, or the fraction of posterior trees that contain the clade) to generate a total score for the tree. This would generate a Maximum credibility tree. However, both methods are used in various contexts.^[2]

References

1. ^ Drummond, Alexei. "Summarizing posterior trees". http://beast.bio.ed.ac.uk/Summarizing_posterior_trees. Retrieved 25 October 2010. 
2. ^ Dunn, M.; Greenhill, S. J.; Levinson, S. C.; Gray, R. D. (2011). "Evolved structure of language shows lineage-specific trends in word-order universals". Nature 473 (7345). doi:10.1038/nature09923.  edit