HLAB27Pred: SVM-based precise method for predicting HLA-B*2705 binding peptides in antigenic sequences

Abstract

Binding of Major histocompatibility complex (MHC) peptide is a prerequisite for T cell activation in the immune system. MHC-binding peptides have shown promising results for immunodiagnostics and immunotherapeutic purposes. HLA-B*2705 is found to be associated with the development of variety of autoimmune diseases including Ankylosing spondylitis. Detecting MHC class I allele HLA-B*2705 binding peptides can reduce the number of peptides that need to be tested experimentally. This work describes the implementation of SVM algorithm, developed for the identification of HLA-B*2705 binding peptides in antigenic sequences. The specificity and sensitivity obtained during the development of this server are 85 and 86 %, respectively. Whereas average precision and average recall values were observed to be 85 and 86 %, respectively. Training on wide-scale data made this method more accurate and robust than all available other methods for the HLA-B*2705 allele and would prove its usability in the biomedical domain. A web server HLA-B27pred is available at http://www.nuccore.org/hlab27pred for academic and research purpose.