An in-house perl script using WU-BLAST 2.0 [Gish, W. (1996-2004) http://blast.wustl.edu] was used to annotate predicted genes with Kegg Orthology (KO) identifiers. The protein sequence of each gene was mapped against the Kegg genes database [PMID:22080510] version 68.0 that had been filtered to only include subjects that were annotated with KO identifiers.
