OrthoMCL: identification of ortholog groups for eukaryotic genomes.

The paper introduces OrthoMCL, a scalable method for constructing groups of orthologous genes across multiple eukaryotic taxa. Identifying such groups is useful for genome annotation, studies of gene and protein evolution, comparative genomics, and finding taxonomically restricted sequences, but methods that work for prokaryotes are difficult to apply to eukaryotes because larger genomes contain multiple paralogs and sequence information is often incomplete. OrthoMCL uses a Markov Cluster algorithm to group putative orthologs and paralogs, performing similarly to the INPARANOID algorithm on two genomes while extending to cluster orthologs from many species.

OrthoMCL clusters are coherent with groups identified by EGO, and improved recognition of recent paralogs lets overlapping EGO groups representing the same gene be merged. Comparison with previously assigned EC annotations suggests a high degree of reliability, implying utility for automated eukaryotic genome annotation. The method was applied to proteome data from seven genomes (human, fly, worm, yeast, Arabidopsis, the malaria parasite Plasmodium falciparum, and E. coli), is queryable through a web interface, and its analysis of P. falciparum clusters identified enzymes that had been incompletely annotated.