The growth of genomic and proteomic databases has led to an enormous annotation gap in which the amount of uncharacterised protein sequences has dramatically increased relative to the ability of slow and costly experimental ...