Prediction of protein-protein interface residues using sequence neighborhood and surface properties

Authors:
Yasir Arafat;Joarder Kamruzzaman;Gour Karmakar
Affiliations:
Faculty of Information Technology, Monash University, Australia;Faculty of Information Technology, Monash University, Australia;Faculty of Information Technology, Monash University, Australia
Venue:
ISNN'06 Proceedings of the Third international conference on Advances in Neural Networks - Volume Part III
Year:
2006

Citing 3
Cited 0

The nature of statistical learning theory

The nature of statistical learning theory
Improved prediction of protein--protein binding sites using a support vector machines approach

Bioinformatics
A comparison of methods for multiclass support vector machines

IEEE Transactions on Neural Networks

Quantified Score

Hi-index	0.00

Visualization

Abstract

An increasing number of protein structures with unknown functions have been solved in the recent years. But understanding the mechanism of protein-protein association still remains one of the biggest problems in structural bioinformatics. Significant research efforts have been dedicated to the identification of protein binding sites and detecting specific amino acid residues which have important connotations ranging from rational drug design to analysis of metabolic and signal transduction networks. In this paper, we present a support vector machine (SVM) based model to predict interface residues from amino acid sequences using sequence neighborhood and surface properties. Experiments with a number of surface properties reveal that the prediction accuracy enhances when residue interface propensity and coil interface propensity of amino acid residues are incorporated in the prediction model which is an improvement over the previously proposed model based on sequence neighborhood only. We also analyzed the effectiveness of a recently proposed coding scheme [1] of secondary structures on the proposed model.