Improving disulfide connectivity prediction with sequential distance between oxidized cysteines

Authors:
Chi-Hung Tsai;Bo-Juen Chen;Chen-Hsiung Chan;Hsuan-Liang Liu;Cheng-Yan Kao
Affiliations:
Department of Computer Science and Information Engineering, National Taiwan University Taipei, Taiwan 106;Department of Computer Science and Information Engineering, National Taiwan University Taipei, Taiwan 106;Department of Computer Science and Information Engineering, National Taiwan University Taipei, Taiwan 106;Department of Chemical Engineering and Graduate Institute of Biotechnology, National Taipei University of Technology Taipei, Taiwan 10608;Department of Computer Science and Information Engineering, National Taiwan University Taipei, Taiwan 106
Venue:
Bioinformatics
Year:
2005

Citing 0
Cited 5

Prediction of the disulphide bridges in proteins using SVM

International Journal of Bioinformatics Research and Applications
Disulfide bonding pattern prediction using support vector machine with parameters tuned by multiple trajectory search

AIC'09 Proceedings of the 9th WSEAS international conference on Applied informatics and communications
Prediction of disulfide bonding pattern based on support vector machine with parameters tuned by multiple trajectory search

WSEAS Transactions on Computers
Prediction of disulfide bonding pattern based on a support vector machine and multiple trajectory search

Information Sciences: an International Journal
Disulfide connectivity prediction based on structural information without a prior knowledge of the bonding state of cysteines

Computers in Biology and Medicine

Quantified Score

Hi-index	3.85

Visualization

Abstract

Summary: Predicting disulfide connectivity precisely helps towards the solution of protein structure prediction. In this study, a descriptor derived from the sequential distance between oxidized cysteines (denoted as DOC) is proposed. An approach using support vector machine (SVM) method based on weighted graph matching was further developed to predict the disulfide connectivity pattern in proteins. When DOC was applied, prediction accuracy of 63% for our SVM models could be achieved, which is significantly higher than those obtained from previous approaches. The results show that using the non-local descriptor DOC coupled with local sequence profiles significantly improves the prediction accuracy. These improvements demonstrate that DOC, with a proper scaling scheme, is an effective feature for the prediction of disulfide connectivity. The method developed in this work is available at the web server PreCys (prediction of cys--cys linkages of proteins). Availability: http://bioinfo.csie.ntu.edu.tw:5433/Disulfide/ Contact: cykao@csie.ntu.edu.tw Supplementary information: Supplementary data, detailed results, tables and information are available at http://bioinfo.csie.ntu.edu.tw:5433/Disulfide/