Use of genetic algorithms and solvation potential to study peptides structure

In this work, genetic algorithms concepts along with a rotamer library for proteins side chains and implicit solvation potential are used to optimize the tertiary structure of peptides. We starting from the known PDB structure of its backbone which is kept fixed while the side chains allowed adopting the conformations present in the rotamer library. It was used rotamer library independent of backbone and a implicit solvation potential. The structure of Mastoporan-X was predicted using several force fields with a growing complexity; we started it with a field where the only present interaction was Lennard-Jones. We added the Coulombian term and we considered the solvation effects through a term proportional to the solvent accessible area. This paper present good and interesting results obtained using the potential with solvation term and rotamer library. Hence, the algorithm (called YODA) presented here can be a good tool to the prediction problem. (c) 2007 Elsevier B.V. All rights reserved.