Estimating large phylogenies using "supertree" techniques

Numerous issues in coral-reef biology would benefit from the application of a formal comparative evolutionary approach. Hence, estimating a phylogeny of Scleractinia (over 1314 extant species) is a growing problem in the field. However, an important constraint on estimating large phylogenies is computational effort, which scales (super-)exponentially with the size of a problem (as taxa) and model complexity. Another important limitation is data acquisition, because no research group has access to all faunas. Currently, the most promising tack around these issues is to use a "supertree" algorithm that combines subtrees previously generated from smaller, overlapping datasets (Bininda-Emonds 2004 Trends Ecol Evol 19 315). We used the supertree algorithm matrix representation with parsimony to combine existing hypotheses of coral relationships and provide the most comprehensive species-level estimate of scleractinian phylogeny, comprised of 358 species (27% of extant species), 141 genera (63%) and 23 families (92%) from all seven suborders. Conservative grafting of species at the base of congeneric groups with uncontroversial monophyletic status resulted in a much larger, though less resolved tree of 1016 taxa, potentially useful in formal comparative studies of character evolution. The tree has resolved several unanticipated large clades united by interesting biological features that we are just beginning to explore: 1) A large clade of potentially 100 Atlantic species from five different families, all with protruding corallites. 2) The evolution of hermaphroditism appears to evolve in concert with the acquisition of brooding. 3) A large clade of three familes comprised of solitary, deepwater species. Below is the the coral "meta-tree" (J Pandolfi, pers comm) as an Adam's consensus. The left and right sister subclades correspond, respectively, to Ramono and Palumbi's (1996 Science 271: 640) "complex" and "robust" subclades.