Prediction of Contact Numbers of Amino Acid Residues Using a Neural Network Model

Abstract

The number of contacts is an important characteristic of amino acid residues in proteins. This characteristic relates to the accessibility to solvent and may be used in the prediction of the protein contact matrices. Here, we propose an approach to the prediction of the number of residue contacts in proteins on the basis of protein sequence, position specific scoring matrices for homologous sequences, and neural network algorithm.