An Enhanced Genetic Algorithm for Protein Structure Prediction Using the 2D Hydrophobic-Polar Model

This paper presents an enhanced genetic algorithm for the protein structure prediction problem. A new fitness function, that uses the concept of radius of gyration, is proposed. Also, a novel operator called partial optimization, together with different strategies for performance improvement, are described. Tests were done with five different amino acid chains from 20 to 85 residues long and better results were obtained, when compared with those in the current literature. Results are promising and suggest the suitability of the proposed method for protein structure prediction using the 2D HP model. Further experiments shall be done with longer amino acid chains as well as with realworld proteins.

• Type: Conference
• Year: 2005
• Where: AE
• Authors: Heitor S. Lopes, Marcos P. Scapin
• Keywords: AE 2005, Amino Acid Chains, Protein Structure Prediction, Structure Prediction Problem