A large peptidome dataset improves HLA class I epitope prediction across most of the human population

• Published in: Nature biotechnology Vol. 38; no. 2; pp. 199 - 209
• Main Authors: Sarkizova, Siranush, Klaeger, Susan, Le, Phuong M., Li, Letitia W., Oliveira, Giacomo, Keshishian, Hasmik, Hartigan, Christina R., Zhang, Wandi, Braun, David A., Ligon, Keith L., Bachireddy, Pavan, Zervantonakis, Ioannis K., Rosenbluth, Jennifer M., Ouspenskaia, Tamara, Law, Travis, Justesen, Sune, Stevens, Jonathan, Lane, William J., Eisenhaure, Thomas, Lan Zhang, Guang, Clauser, Karl R., Hacohen, Nir, Carr, Steven A., Wu, Catherine J., Keskin, Derin B.
• Format: Journal Article
• Language: English
• Published: New York Nature Publishing Group US 01.02.2020; Nature Publishing Group
• Subjects: 631/114/2397; 631/250/21; 631/250/580; 631/45/611; 692/308/575; Agriculture; Algorithms; Alleles; Amino Acid Motifs; Analysis; Antigenic determinants; Bioinformatics; Biomedical and Life Sciences; Biomedical Engineering/Biotechnology; Biomedicine; Biotechnology; Cancer immunotherapy; Cancer vaccines; Cell Line; Databases, Protein; Datasets; Epitopes; Epitopes - metabolism; Genetic Loci; Histocompatibility antigen HLA; Histocompatibility antigens; Histocompatibility Antigens Class I - metabolism; HLA histocompatibility antigens; Human populations; Humans; Immunotherapy; Life Sciences; Ligands; Mass spectrometry; Mass spectroscopy; Peptide Hydrolases - metabolism; Peptides; Peptides - chemistry; Peptides - metabolism; Prediction models; Predictions; Proteasome Endopeptidase Complex - metabolism; Proteome - metabolism; Transcription; Tumor cell lines; Vaccines; Massachusetts
• ISSN: 1087-0156; 1546-1696
• DOI: 10.1038/s41587-019-0322-9

Prediction of HLA epitopes is important for the development of cancer immunotherapies and vaccines. However, current prediction algorithms have limited predictive power, in part because they were not trained on high-quality epitope datasets covering a broad range of HLA alleles. To enable prediction of endogenous HLA class I-associated peptides across a large fraction of the human population, we used mass spectrometry to profile >185,000 peptides eluted from 95 HLA-A, -B, -C and -G mono-allelic cell lines. We identified canonical peptide motifs per HLA allele, unique and shared binding submotifs across alleles and distinct motifs associated with different peptide lengths. By integrating these data with transcript abundance and peptide processing, we developed HLAthena, providing allele-and-length-specific and pan-allele-pan-length prediction models for endogenous peptide presentation. These models predicted endogenous HLA class I-associated ligands with 1.5-fold improvement in positive predictive value compared with existing tools and correctly identified >75% of HLA-bound peptides that were observed experimentally in 11 patient-derived tumor cell lines. Prediction of HLA class I epitopes is improved in accuracy and breath with peptidomes from 95 mono-allelic cell lines.