Nanotech 2002 Vol. 2
Nanotech 2002 Vol. 2
Technical Proceedings of the 2002 International Conference on Computational Nanoscience and Nanotechnology

Sequence and Biological Structure Chapter 2

Solving the Protein Structure Prediction Problem Through a Multiobjective Genetic Algorithm

Authors: R.O. Day, J.B. Zydallis, G.B. Lamont and R. Pachter

Affilation: Air Force Institute of Technology, United States

Pages: 32 - 35

Keywords: energy minimization, protein structure prediction problem, genetic algorithm, fast messy genetic algorithm

The Protein Structure Prediction (PSP) problem is a Grand Challenge problem among biochemists, computer scientists and engineers alike. Solving this problem involves correctly predicting the geometrical conformation of a fully folded protein. This paper focuses on CHARMm energy minimization and the use of a genetic algorithm, the fast messy genetic algorithm (fmGA), to obtain solutions to this optimization problem. The fmGA is a novel algorithm that explicitly manipulates building blocks (BBs) in order to obtain "good" solutions to an optimization problem. In order to obtain these "good" solutions, fully speci¯ed competitive templates are used within the fmGA to evaluate the BBs found. This paper presents "good" results of an analysis of various competitive template schemes for the application of the fmGA to the PSP of [Met]-Enkephelin and the much larger Polyalanine peptide.

ISBN: 0-9708275-6-3
Pages: 504