Format

Send to

Choose Destination
Bioinformatics. 2005 Dec 15;21(24):4416-9. Epub 2005 Oct 13.

Improving disulfide connectivity prediction with sequential distance between oxidized cysteines.

Author information

1
Department of Computer Science and Information Engineering, National Taiwan University, Taipei, Taiwan 106.

Abstract

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).

PMID:
16223789
DOI:
10.1093/bioinformatics/bti715
[Indexed for MEDLINE]

Supplemental Content

Full text links

Icon for Silverchair Information Systems
Loading ...
Support Center