Improving consensus contact prediction via server correlation reduction

Xin Gao,Jinbo Xu,Dongbo Bu,Ming Li

doi:10.1186/1472-6807-9-28

Abstract

BackgroundProtein inter-residue contacts play a crucial role in the determination and prediction of protein structures. Previous studies on contact prediction indicate that although template-based consensus methods outperform sequence-based methods on targets with typical templates, such consensus methods perform poorly on new fold targets. However, we find out that even for new fold targets, the models generated by threading programs can contain many true contacts. The challenge is how to identify them.ResultsIn this paper, we develop an integer linear programming model for consensus contact prediction. In contrast to the simple majority voting method assuming that all the individual servers are equally important and independent, the newly developed method evaluates their correlation by using maximum likelihood estimation and extracts independent latent servers from them by using principal component analysis. An integer linear programming method is then applied to assign a weight to each latent server to maximize the difference between true contacts and false ones. The proposed method is tested on the CASP7 data set. If the top L/5 predicted contacts are evaluated where L is the protein size, the average accuracy is 73%, which is much higher than that of any previously reported study. Moreover, if only the 15 new fold CASP7 targets are considered, our method achieves an average accuracy of 37%, which is much better than that of the majority voting method, SVM-LOMETS, SVM-SEQ, and SAM-T06. These methods demonstrate an average accuracy of 13.0%, 10.8%, 25.8% and 21.2%, respectively.ConclusionReducing server correlation and optimally combining independent latent servers show a significant improvement over the traditional consensus methods. This approach can hopefully provide a powerful tool for protein structure refinement and prediction use.

Highlights

Protein inter-residue contacts play a crucial role in the determination and prediction of protein structures
Data Set Server Selection To evaluate the performance of the proposed consensus method, six threading-based protein structure prediction servers are used: FOLDpro [44], mGenThreader [45,46], RAPTOR [47,48], FUGUE3 [38], SAM-T02 [41], and SPARK3 [42]
All server models are downloaded from the CASP7 website, except for mGenThreader, which does not participate in CASP7

Summary

Introduction

Protein inter-residue contacts play a crucial role in the determination and prediction of protein structures. Misura et al [9] have revised the widely-used ab initio folding program, Rosetta [10], by incorporating inter-residue contact information as a component of Rosetta's energy function, and shown that the revised Rosetta exhibits a better computational efficiency, and a better prediction accuracy. One of the major advantages of these two programs over the others is that both depend on contacts and distance restraints, extracted from multiple templates, to refine the template-based models. It has been shown by Zhang et al that ab initio prediction methods can benefit from contact predictions with an accuracy that is higher than 22% [13]

Objectives

Methods

Results

Discussion

Conclusion