Combing genetic algorithm with neural network technique for protein inter-residue spatial distance prediction

The spatial distance of amino acids in a protein sequence is one of the important factors, which determine the three-dimension structure (tertiary structure). In this paper, we describe a genetic algorithm (GA) based radial basis function neural networks (RBFNN), whose hidden centers and radial basis function widths is optimized by the GA, to learn how primary structure (residue sequence) affects the spatial proximity of the amino acids in the soybean protein sequences and predict the residues spatial distance in three-dimensional space. Experimental results indicate that the proposed network has a good performance in soybean protein sequences residue spatial distance prediction.