Prediction of human major histocompatibility complex class II binding peptides by continuous kernel discrimination method

• Published in: Artificial intelligence in medicine Vol. 55; no. 2; pp. 107 - 115
• Main Authors: He, Ju, Yang, Guobing, Rao, Hanbing, Li, Zerong, Ding, Xianping, Chen, Yuzong
• Format: Journal Article
• Language: English
• Published: Netherlands Elsevier B.V 01.06.2012
• Subjects: Area Under Curve; Artificial Intelligence; Benchmarking; Binding; Binding Sites; Computer Simulation; Continuous kernel discrimination; Databases, Protein; Feature selection; Histocompatibility Antigens Class II - chemistry; Humans; Internal Medicine; Kernels; Machine learning; Major histocompatibility complex class II peptides; Mathematical models; Metropolis Monte Carlo simulated annealing; Models, Chemical; Monte Carlo Method; Monte Carlo methods; Other; Peptides; Peptides - chemistry; Protein Binding; Feature selection; Major histocompatibility complex class II peptides; Metropolis Monte Carlo simulated annealing; Continuous kernel discrimination
• DOI: 10.1016/j.artmed.2011.10.005
• ISSN: 0933-3657; 1873-2860

Accurate prediction of major histocompatibility complex (MHC) class II binding peptides helps reducing the experimental cost for identifying helper T cell epitopes, which has been a challenging problem partly because of the variable length of the binding peptides. This work is to develop an accurate model for predicting MHC-binding peptides using machine learning methods. In this work, a machine learning method, continuous kernel discrimination (CKD), was used for predicting MHC class II binders of variable lengths. The composition transition and distribution features were used for encoding peptide sequence and the Metropolis Monte Carlo simulated annealing approach was used for feature selection. Feature selection was found to significantly improve the performance of the model. For benchmark dataset Dataset-1, the number of features is reduced from 147 to 24 and the area under the receiver operating characteristic curve (AUC) is improved from 0.8088 to 0.9034, while for benchmark dataset Dataset-2, the number of features is reduced from 147 to 44 and the AUC is improved from 0.7349 to 0.8499. An optimal CKD model was derived from the feature selection and bandwidth optimization using 10-fold cross-validation. Its AUC values are between 0.831 and 0.980 evaluated on benchmark datasets BM-Set1 and are between 0.806 and 0.949 on benchmark datasets BM-Set2 for MHC class II alleles. These results indicate a significantly better performance for our CKD model over other earlier models based on the training and testing of the same datasets. Our study suggested that the CKD method outperforms other machine learning methods proposed earlier in the prediction of MHC class II biding peptides. Moreover, the choice of the cut-off for CKD classifier is crucial for its performance.