PREDICTION OF MHC CLASS-II BINDING PEPTIDES BY SUPPORT VECTOR MACHINES
利用支持向量机预测II类MHC分子结合多肽

The binding of antigenic peptide sequences to major histocompatibility complex(MHC) molecules is a prerequisite for stimulation of cytotoxic T cell responses. Prediction methods for identifying binding peptides could minimize the number of peptides required to be synthesized and assayed, therefore reduce laboratory time and costs. Support Vector Machines (SVM) have been used to predict the binding capacity of polypeptides to MHC class-II molecules encoded by the gene HLA-DR4(B1*0401). 650 peptide sequences whose binding capacity were known were separated into two groups-training set and testing set in the proportion of 3 to 1. Then they were implemented by SVM. About 77.62% of Binding peptide sequences and 71.79% of Non-binding peptide sequences were well classified. This illustrates that the application of SVM to the identification of potential immunotherapeutic peptides is feasible.