Machine learning predictions of MHC-II specificities reveal alternative binding mode of class II epitopes

• DOI: 10.1016/j.immuni.2023.03.009
• Publication Date: 2023-04-05T21:46:25Z
• Authors: Julien Racle; Philippe Guillaume; Julien Schmidt; Justine Michaux; Amédé Larabi; Kelvin Lau; Marta A.S. Perez; Giancarlo Croce; Raphaël Genolet; George Coukos; Vincent Zoete; Florence Pojer; Michal Bassani-St...; Alexandre Harari; David Gfeller
• Keywords: 0301 basic medicine; immunopeptidome; Immunology; Humans; Animals; Mice; Cattle; Epitopes, T-Lymphocyte; Ligands; Peptides; Protein Binding; Chickens/metabolism; Machine Learning; Histocompatibility Antigens Class II; Alleles; MHC-II binding motifs; MHC-II ligand binding modes; antigen presentation; class II epitope predictions; computational immunology; immunopeptidomics; reverse binding mode; motifs; Epitopes, T-Lymphocyte; Ligands; Machine Learning; Mice; 03 medical and health sciences; proteomics; Immunology and Allergy; Humans; Animals; Alleles; Histocompatibility Antigens Class II; hla-dr; cell; 3. Good health; antigen presentation; Infectious Diseases; identification; Cattle; recognition; protein; Peptides; peptide binding; Chickens; Protein Binding

CD4+ T cells orchestrate the adaptive immune response against pathogens and cancer by recognizing epitopes presented on class II major histocompatibility complex (MHC-II) molecules. The high polymorphism of MHC-II genes represents an important hurdle toward accurate prediction identification cell epitopes. Here we collected curated a dataset 627,013 unique ligands identified mass spectrometry. This enabled us to precisely determine binding motifs 88 alleles across humans, mice, cattle, chickens. Analysis these specificities combined with X-ray crystallography refined our understanding molecular determinants revealed widespread reverse-binding mode in HLA-DP ligands. We then developed machine-learning framework accurately predict any allele. tool improves expands predictions enables discover viral bacterial following aforementioned mode.