Leveraging Information Across HLA Alleles/Supertypes Improves Epitope Prediction

We present a model for predicting HLA class I restricted CTL epitopes. In contrast to almost all other work in this area, we train a single model on epitopes from all HLA alleles and supertypes, yet retain the ability to make epitope predictions for specific HLA alleles. We are therefore able to leverage data across all HLA alleles and/or their supertypes, automatically learning what information should be shared and also how to combine allele-specific, supertype-specific, and global information in a principled way. We show that this leveraging can improve prediction of epitopes having HLA alleles with known supertypes, and dramatically increases our ability to predict epitopes having alleles which do not fall into any of the known supertypes. Our model, which is based on logistic regression, is simple to implement and understand, is solved by finding a single global maximum, and is more accurate (to our knowledge) than any other model.

[1]  Arne Elofsson,et al.  Prediction of MHC class I binding peptides, using SVMHC , 2002, BMC Bioinformatics.

[2]  Lei Wang,et al.  Analysis of therapeutic effects of soothing the liver and regulating the stomach on 80 cases of functional dyspepsia , 1996 .

[3]  Yingdong Zhao,et al.  Application of support vector machines for T-cell epitopes prediction , 2003, Bioinform..

[4]  S Brunak,et al.  Sensitive quantitative predictions of peptide-MHC binding by a 'Query by Committee' artificial neural network approach. , 2003, Tissue antigens.

[5]  Hai-Long Dong,et al.  Prediction of HLA-A2-restricted CTL epitope specific to HCC by SYFPEITHI combined with polynomial method. , 2005, World journal of gastroenterology.

[6]  Gajendra P. S. Raghava,et al.  MHCBN: a comprehensive database of MHC binding and non-binding peptides , 2003, Bioinform..

[7]  O. Lund,et al.  novel sequence representations Reliable prediction of T-cell epitopes using neural networks with , 2003 .

[8]  Gajendra P. S. Raghava,et al.  SVM based method for predicting HLA-DRB1*0401 binding peptides in an antigen sequence , 2004, Bioinform..

[9]  E. Rosenberg,et al.  Rapid Definition of Five Novel HLA-A∗3002-Restricted Human Immunodeficiency Virus-Specific Cytotoxic T-Lymphocyte Epitopes by Elispot and Intracellular Cytokine Staining Assays , 2001, Journal of Virology.

[10]  Gajendra P.S. Raghava,et al.  Prediction of CTL epitopes using QM, SVM and ANN techniques. , 2004, Vaccine.

[11]  A. McMichael,et al.  Opinion — vaccines: The quest for an AIDS vaccine: is the CD8+ T-cell approach feasible? , 2002, Nature Reviews Immunology.

[12]  J. Skolnick,et al.  Application of an artificial neural network to predict specific class I MHC binding peptide sequences , 1998, Nature Biotechnology.

[13]  O. Lund,et al.  An integrative approach to CTL epitope prediction: A combined algorithm integrating MHC class I binding, TAP transport efficiency, and proteasomal cleavage predictions , 2005, European journal of immunology.

[14]  H. Rammensee,et al.  SYFPEITHI: database for MHC ligands and peptide motifs , 1999, Immunogenetics.