Title: Predicting contact map using Radial Basis Function Neural Network with Conformational Energy Function

Authors: Peng Chen, De-Shuang Huang, Xing-Ming Zhao, Xueling Li

Addresses: Intelligent Computing Lab, Hefei Institute of Intelligent Machines, Chinese Academy of Sciences, Hefei, Anhui, 230031, China; Department of Automation, University of Science and Technology of China, Hefei, Anhui, 230026, China. ' Intelligent Computing Lab, Hefei Institute of Intelligent Machines, Chinese Academy of Sciences, Hefei, Anhui, 230031, China. ' Intelligent Computing Lab, Hefei Institute of Intelligent Machines, Chinese Academy of Sciences, Hefei, Anhui, 230031, China. ' Intelligent Computing Lab, Hefei Institute of Intelligent Machines, Chinese Academy of Sciences, Hefei, Anhui, 230031, China

Abstract: Contact map, which is important to understand and reconstruct protein|s three-dimensional (3D) structure, may be helpful to solve the protein|s 3D structure. This paper presents a novel approach to predict the contact map using Radial Basis Function Neural Network (RBFNN) optimised by Conformational Energy Function (CEF) based on chemico-physical knowledge of amino acids. Finally, the results are trimmed by Short-Range Contact Function (SRCF). Consequently, it can be found that our proposed method is better than the existing methods such as PROFcon and the PE-based method. Particularly, this method can accurately predict 35% of contacts at a distance cutoff of 8 Å.

Keywords: contact maps; radial basis function NNs; neural networks; RBFNN; conformational energy function; CEF; principal component analysis; PCA; short-range contact function; SRCF; bioinformatics; protein structure; amino acids.

DOI: 10.1504/IJBRA.2008.018340

International Journal of Bioinformatics Research and Applications, 2008 Vol.4 No.2, pp.123 - 136

Published online: 17 May 2008 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article