Article: Prediction of HLA-DRB1*0401 binding peptides using support vector machine Journal: International Journal of Data Mining and Bioinformatics (IJDMB) 2014 Vol.10 No.2 pp.189 - 205 Abstract: In recent years, many machine learning methods have been developed to predict HLA binding peptides. However, because only limited types of descriptors characterising the protein features are included in these approaches, these methods have poor prediction accuracy. In this study, we applied support vector machine methods to predict the peptides that bind to the major histocompatibility complexes Class II molecule HLA-DRB1*0401 using six sets of molecular descriptors characterising the primary structures of the peptides. We found that some feature groups provided good prediction accuracies and the overall accuracies were greater than 95% and some feature groups had poor accuracies of only 50%. The performance was improved significantly by additional feature selection and the overall accuracies from each group or combination of descriptors were greater than 90%. Of note, the inclusion of necessary informative and discriminative descriptors improved the prediction accuracies. Inderscience Publishers - linking academia, business and industry through research

Title: Prediction of HLA-DRB1*0401 binding peptides using support vector machine

Authors: Wenli Huang; Guobing Yang; Xiaojun Zhao; Zerong Li

Addresses: Institute for Nanobiomedical Technology and Membrane Biology, West China Hospital, Sichuan University, Chengdu 610041, Sichuan, China; Regenerative Medicine Research Center, West China Hospital, Sichuan University, Chengdu 610041, Sichuan, China ' College of Chemical Engineering, Sichuan University, Chengdu 610065, China ' Institute for Nanobiomedical Technology and Membrane Biology, West China Hospital, Sichuan University, Chengdu 610041, Sichuan, China ' College of Chemistry, Sichuan University, Chengdu 610065, China

Abstract: In recent years, many machine learning methods have been developed to predict HLA binding peptides. However, because only limited types of descriptors characterising the protein features are included in these approaches, these methods have poor prediction accuracy. In this study, we applied support vector machine methods to predict the peptides that bind to the major histocompatibility complexes Class II molecule HLA-DRB1*0401 using six sets of molecular descriptors characterising the primary structures of the peptides. We found that some feature groups provided good prediction accuracies and the overall accuracies were greater than 95% and some feature groups had poor accuracies of only 50%. The performance was improved significantly by additional feature selection and the overall accuracies from each group or combination of descriptors were greater than 90%. Of note, the inclusion of necessary informative and discriminative descriptors improved the prediction accuracies.

Keywords: HLA-DRB1*0401; MHC binding peptides; molecular descriptors; feature selection; support vector machines; SVM; prediction accuracy; bioinformatics.

DOI: 10.1504/IJDMB.2014.064015

International Journal of Data Mining and Bioinformatics, 2014 Vol.10 No.2, pp.189 - 205

Received: 30 Jun 2011
Accepted: 07 Jul 2012
Published online: 21 Oct 2014 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article

Title: Prediction of HLA-DRB1*0401 binding peptides using support vector machine

Keep up-to-date