Abstract
Orphan enzymes are the enzymes whose catalytic activities are described whereas the gene sequences are unknown. For the aim to address the genes, which should be encoded in the genome, corresponding to those orphan enzymes, we explored the space of hypothetical proteins predicted in the human genome. To construct HMM profiles of functionally similar but weakly homologous sequences were prepared so that minimal structure for the function can be identified. Among hypothetical proteins in H-invDB_7.0 and enzymes in DDBJ, we successfully addressed 287 hypothetical proteins to specific EC numbers. We assume that the profiles suggested active sites of low similarities.