Abstract
Abstract Realizing how proteins work has been eagerly studied to discover the secret of human body in post genome era. Several well-development tools such as X-ray crystallography, NMR spectroscopy, or homology modeling etc., analyzing protein structure and function have limitation. Therefore, in this paper we propose a heuristic algorithm that can suggest the possible functional residues for studying further. At first, we choose rat imidase as our interesting target protein and imdase superfamily as assistant data. Then, we classify the imidase superfamily into distinct groups according to its function. Moreover, each group aligned uses cross-reference skill to vote residues of rat imidase. Eventually, higher scores correspond to higher possible critical residue candidates. In this study, we predict the critical residue candidates of rat imidase, and reduce the number of residues from 519 to 98. This is helpful to biologists, and they can save their time, money, labor and experimental times. This is the contribution of our algorithm.