MHC-NP: Predicting peptides naturally processed by the MHC

• Published in: Journal of immunological methods 2013-12, Vol.400-401, p.30-36
• Main Authors: Giguère, Sébastien, Drouin, Alexandre, Lacoste, Alexandre, Marchand, Mario, Corbeil, Jacques, Laviolette, François
• Format: Article
• Language: English
• Subjects: Algorithms; Animals; Antigen Presentation; Artificial Intelligence; Epitope; Epitope Mapping - methods; epitopes; H-2 Antigens - chemistry; H-2 Antigens - immunology; Histocompatibility Antigen H-2D - chemistry; Histocompatibility Antigen H-2D - immunology; HLA-A2 Antigen - chemistry; HLA-A2 Antigen - immunology; HLA-B Antigens - chemistry; HLA-B Antigens - immunology; Humans; Immunology; Kernel; Machine learning; Major Histocompatibility Complex - immunology; MHC; Mice; motivation; peptides; Peptides - chemistry; Peptides - immunology; prediction; Protein Binding; Software; Vaccines; Vaccinology
• ISSN: 0022-1759; 1872-7905
• DOI: 10.1016/j.jim.2013.10.003

We present MHC-NP, a tool for predicting peptides naturally processed by the MHC pathway. The method was part of the 2nd Machine Learning Competition in Immunology and yielded state-of-the-art accuracy for the prediction of peptides eluted from human HLA-A*02:01, HLA-B*07:02, HLA-B*35:01, HLA-B*44:03, HLA-B*53:01, HLA-B*57:01 and mouse H2-Db and H2-Kb MHC molecules. We briefly explain the theory and motivations that have led to developing this tool. General applicability in the field of immunology and specifically epitope-based vaccine are expected. Our tool is freely available online and hosted by the Immune Epitope Database at http://tools.immuneepitope.org/mhcnp/. •We propose a tool to predict peptides naturally processed by MHC pathway.•Our method outperforms state-of-the-art methods based on MHC-peptide binding affinity.•In the 2012 MLI competition, our tool was the most accurate on 4 out of 8 alleles.•Broad applications in the field of immunology and epitope vaccine design are expected.