A congruence index for testing topological similarity between trees

Abstract

Motivation: Phylogenetic trees are omnipresent in evolutionary biology and the comparison of trees plays a central role there. Tree congruence statistics are based on the null hypothesis that two given trees are not more congruent (topologically similar) than expected by chance. Usually, one searches for the most parsimonious evolutionary scenario relating two trees and then one tests the null hypothesis by generating a high number of random trees and comparing these to the one between the observed trees. However, this approach requires a lot of computational work (human and machine) and the results depend on the evolutionary assumptions made.

Results: We propose an index, I(cong), for testing the topological congruence between trees with any number of leaves, based on maximum agreement subtrees (MAST). This index is straightforward, simple to use, does not rely on parametrizing the likelihood of evolutionary events, and provides an associated confidence level.

Availability: A web site has been created that allows rapid and easy online computation of this index and of the associated P-value at http://www.ese.u-psud.fr/bases/upresa/pages/devienne/index.html