Protein structure prediction remains one of the most important challenges in molecular biology. Contact maps have been extensively used as a simplified representation of protein structures. In this work, we propose a multi-objective evolutionary approach for contact map prediction. The proposed method bases the prediction on a set of physico-chemical properties and structural features of the amino acids, as well as evolutionary information in the form of an amino acid position specific scoring matrix (PSSM). The proposed technique produces a set of decision rules that identify contacts between amino acids. Results obtained by our approach are presented and confirm the validity of our proposal.
展开▼