Protein binding site prediction by combining hidden Markov support vector machine and profile-based propensities.

Bin Liu,Xiaolong Wang,Fule Liu,Bingquan Liu

doi:10.1155/2014/464093

Abstract

Identification of protein binding sites is critical for studying the function of the proteins. In this paper, we proposed a method for protein binding site prediction, which combined the order profile propensities and hidden Markov support vector machine (HM-SVM). This method employed the sequential labeling technique to the field of protein binding site prediction. The input features of HM-SVM include the profile-based propensities, the Position-Specific Score Matrix (PSSM), and Accessible Surface Area (ASA). When tested on different data sets, the proposed method showed promising results, and outperformed some closely relative methods by more than 10% in terms of AUC.

Highlights

Prediction of protein binding sites provides valuable information for studying the function of proteins
We proposed a computational method for protein binding site prediction by combining the hidden Markov support vector machine and the order profile propensity
The second hidden Markov support vector machine (HM-SVM) predictor using the order profile propensity as an extra feature achieved the best performance on all the six data sets, especially its AUC score being about 10% higher than that of the first HM-SVM predictor, indicating that order profile propensity can significantly improve the performance of the HM-SVM based methods

Summary

Introduction

Prediction of protein binding sites provides valuable information for studying the function of proteins. The most efficient approaches are the computational methods. By using these approaches, the functionally important amino acid residues can be identified [1]. The functionally important amino acid residues can be identified [1] These computational methods used different features extracted from protein sequences, PSSM, or structure information. The conservation scores of amino acid are often used as features, because the protein binding sites are more conserved than other surface residues [4]. One of the most widely used features is the Accessible Surface Area (ASA) [4], because the binding sites show higher ASA values than those of the other surface residues [6]

Methods

Results

Conclusion