Characterization Of The Binding Profile Of Peptide To Transporter Associated With Antigen Processing (TAP) Using Support Vector Machine

The antigen in the specific immunity is divided into endogenous antigen and exogenous antigen,and the exogenous antigen which is produced by other cells(such as bacteria or fungi)and are presented by specific antigen presenting cells.Endogenous antigens mainly refer to that is produce in non full-time antigen presenting cells,such as virus infected cells and tumor cells.The endogenous antigen is decomposed into short peptide of different lengths in the cytoplasm,the peptide fragment is selectively transport to the endoplasmic reticulum by the transporter associated with antigen processing under the ATP energy supply,where they bind to the binding groove of major histocompatibility complex(MHC)class I molecules.and bind to the newly assembled MHC-I molecule in the endoplasmic reticulum.The two conjugates are transported to the cell membrane surface through Golgi body for CD8+T cell recognition.In this work,we have studied the predictive performance of support vector machine models trained on a large dataset consisting of 613 nonamer peptides of known affinity to TAP.Predictive performance of these TAP affinity models was evaluated under 10-fold cross-validation experiments and measured using Pearson's correlation coefficients(Rp).The results show that every peptide position(P1–P9)contributes to TAP binding,although the largest contributions to binding correspond to the C-terminal end and The second is the amino acids on the P1,P2,and P3...