{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,4,15]],"date-time":"2026-04-15T04:54:36Z","timestamp":1776228876062,"version":"3.50.1"},"reference-count":48,"publisher":"Oxford University Press (OUP)","issue":"Supplement_1","license":[{"start":{"date-parts":[[2020,7,13]],"date-time":"2020-07-13T00:00:00Z","timestamp":1594598400000},"content-version":"vor","delay-in-days":12,"URL":"http:\/\/creativecommons.org\/licenses\/by-nc\/4.0\/"}],"funder":[{"name":"Amazon AWS Machine Learning Research Award"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"published-print":{"date-parts":[[2020,7,1]]},"abstract":"<jats:title>Abstract<\/jats:title>\n               <jats:sec>\n                  <jats:title>Motivation<\/jats:title>\n                  <jats:p>Accurate prediction of binding between a major histocompatibility complex (MHC) allele and a peptide plays a major role in the synthesis of personalized cancer vaccines. The immune system struggles to distinguish between a cancerous and a healthy cell. In a patient suffering from cancer who has a particular MHC allele, only those peptides that bind with the MHC allele with high affinity, help the immune system recognize the cancerous cells.<\/jats:p>\n               <\/jats:sec>\n               <jats:sec>\n                  <jats:title>Results<\/jats:title>\n                  <jats:p>MHCAttnNet is a deep neural model that uses an attention mechanism to capture the relevant subsequences of the amino acid sequences of peptides and MHC alleles. It then uses this to accurately predict the MHC-peptide binding. MHCAttnNet achieves an AUC-PRC score of 94.18% with 161 class I MHC alleles, which outperforms the state-of-the-art models for this task. MHCAttnNet also achieves a better F1-score in comparison to the state-of-the-art models while covering a larger number of class II MHC alleles. The attention mechanism used by MHCAttnNet provides a heatmap over the amino acids thus indicating the important subsequences present in the amino acid sequence. This approach also allows us to focus on a much smaller number of relevant trigrams corresponding to the amino acid sequence of an MHC allele, from 9251 possible trigrams to about 258. This significantly reduces the number of amino acid subsequences that need to be clinically tested.<\/jats:p>\n               <\/jats:sec>\n               <jats:sec>\n                  <jats:title>Availability and implementation<\/jats:title>\n                  <jats:p>The data and source code are available at https:\/\/github.com\/gopuvenkat\/MHCAttnNet.<\/jats:p>\n               <\/jats:sec>","DOI":"10.1093\/bioinformatics\/btaa479","type":"journal-article","created":{"date-parts":[[2020,5,5]],"date-time":"2020-05-05T19:19:51Z","timestamp":1588706391000},"page":"i399-i406","source":"Crossref","is-referenced-by-count":53,"title":["MHCAttnNet: predicting MHC-peptide bindings for MHC alleles classes I and II using an attention-based deep neural model"],"prefix":"10.1093","volume":"36","author":[{"given":"Gopalakrishnan","family":"Venkatesh","sequence":"first","affiliation":[{"name":"International Institute of Information Technology Bangalore , Bangalore 560100, India"}]},{"given":"Aayush","family":"Grover","sequence":"additional","affiliation":[{"name":"International Institute of Information Technology Bangalore , Bangalore 560100, India"}]},{"given":"G","family":"Srinivasaraghavan","sequence":"additional","affiliation":[{"name":"International Institute of Information Technology Bangalore , Bangalore 560100, India"}]},{"given":"Shrisha","family":"Rao","sequence":"additional","affiliation":[{"name":"International Institute of Information Technology Bangalore , Bangalore 560100, India"}]}],"member":"286","published-online":{"date-parts":[[2020,7,13]]},"reference":[{"key":"2024021913362065400_btaa479-B1","volume-title":"Cellular and Molecular Immunology E-Book","author":"Abbas","year":"2014"},{"key":"2024021913362065400_btaa479-B2","first-page":"1638","author":"Akbik","year":"2018"},{"key":"2024021913362065400_btaa479-B3","volume-title":"Introduction to Machine Learning","author":"Alpaydin","year":"2014","edition":"3rd edn."},{"key":"2024021913362065400_btaa479-B4","doi-asserted-by":"crossref","first-page":"511","DOI":"10.1093\/bioinformatics\/btv639","article-title":"Gapped sequence alignment using artificial neural networks: application to the MHC class I system","volume":"32","author":"Andreatta","year":"2016","journal-title":"Bioinformatics"},{"key":"2024021913362065400_btaa479-B5","doi-asserted-by":"crossref","first-page":"e0141287","DOI":"10.1371\/journal.pone.0141287","article-title":"Continuous distributed representation of biological sequences for deep proteomics and genomics","volume":"10","author":"Asgari","year":"2015","journal-title":"PLoS One"},{"key":"2024021913362065400_btaa479-B6","author":"Bahdanau","year":"2015"},{"key":"2024021913362065400_btaa479-B7","first-page":"2493","article-title":"Natural language processing (almost) from scratch","volume":"12","author":"Collobert","year":"2011","journal-title":"J. Mach. Learn. Res"},{"key":"2024021913362065400_btaa479-B8","doi-asserted-by":"crossref","first-page":"77","DOI":"10.1177\/2051013614525375","article-title":"MHC class I antigen presentation and implications for developing a new generation of therapeutic vaccines","volume":"2","author":"Comber","year":"2014","journal-title":"Ther. Adv. Vaccines Immunother"},{"key":"2024021913362065400_btaa479-B9","volume-title":"Essential Immunology","author":"Delves","year":"2017"},{"key":"2024021913362065400_btaa479-B10","first-page":"4171","author":"Devlin","year":"2019"},{"key":"2024021913362065400_btaa479-B11","first-page":"1","article-title":"MHC class II binding prediction\u2019a little help from a friend","volume":"2010","author":"Dimitrov","year":"2010","journal-title":"BioMed Res. Int"},{"key":"2024021913362065400_btaa479-B12","doi-asserted-by":"crossref","first-page":"861","DOI":"10.1016\/j.patrec.2005.10.010","article-title":"An introduction to ROC analysis","volume":"27","author":"Fawcett","year":"2006","journal-title":"Pattern Recognit. Lett"},{"key":"2024021913362065400_btaa479-B13","doi-asserted-by":"crossref","first-page":"491","DOI":"10.1016\/0167-5699(93)90264-L","article-title":"Natural history of HLA expression during tumour development","volume":"14","author":"Garrido","year":"1993","journal-title":"Immunol. Today"},{"key":"2024021913362065400_btaa479-B14","volume-title":"Deep Learning","author":"Goodfellow","year":"2016"},{"key":"2024021913362065400_btaa479-B15","doi-asserted-by":"crossref","first-page":"1","DOI":"10.1007\/s00251-008-0341-z","article-title":"NetMHCpan, a method for MHC class I binding prediction beyond humans","volume":"61","author":"Hoof","year":"2009","journal-title":"Immunogenetics"},{"key":"2024021913362065400_btaa479-B16","doi-asserted-by":"crossref","DOI":"10.1101\/468363","article-title":"ACME: pan-specific peptide-MHC class I binding prediction through attention-based deep neural networks","author":"Hu","year":"2018","journal-title":"."},{"key":"2024021913362065400_btaa479-B17","doi-asserted-by":"crossref","first-page":"168","DOI":"10.1038\/nri.2017.131","article-title":"Towards personalized, tumour-specific, therapeutic vaccines for cancer","volume":"18","author":"Hu","year":"2018","journal-title":"Nat. Rev. Immunol"},{"key":"2024021913362065400_btaa479-B18","doi-asserted-by":"crossref","first-page":"394","DOI":"10.1111\/imm.12889","article-title":"Improved methods for predicting peptide binding affinity to MHC class II molecules","volume":"154","author":"Jensen","year":"2018","journal-title":"Immunology"},{"key":"2024021913362065400_btaa479-B19","doi-asserted-by":"crossref","first-page":"3360","DOI":"10.4049\/jimmunol.1700893","article-title":"NetMHCpan-4.0: improved peptide\u2013MHC class I interaction predictions integrating eluted ligand and peptide binding affinity data","volume":"199","author":"Jurtz","year":"2017","journal-title":"J. Immunol"},{"key":"2024021913362065400_btaa479-B20","author":"Kingma","year":"2015"},{"key":"2024021913362065400_btaa479-B21","doi-asserted-by":"crossref","first-page":"1090","DOI":"10.1007\/978-1-4020-5614-7","volume-title":"Encyclopedia of Public Health","author":"Kirch","year":"2008"},{"key":"2024021913362065400_btaa479-B22","doi-asserted-by":"crossref","first-page":"271","DOI":"10.1080\/00207390802213609","article-title":"Pareto 80\/20 law: derivation via random partitioning","volume":"40","author":"Lipovetsky","year":"2009","journal-title":"Int. J. Math. Educ. Sci. Technol"},{"key":"2024021913362065400_btaa479-B23","first-page":"3111","author":"Mikolov","year":"2013"},{"key":"2024021913362065400_btaa479-B24","first-page":"807","author":"Nair","year":"2010"},{"key":"2024021913362065400_btaa479-B25","doi-asserted-by":"crossref","first-page":"33","DOI":"10.1186\/s13073-016-0288-x","article-title":"NetMHCpan-3.0; improved prediction of binding to MHC class I molecules integrating information from multiple receptor and peptide length datasets","volume":"8","author":"Nielsen","year":"2016","journal-title":"Genome Med"},{"key":"2024021913362065400_btaa479-B26","doi-asserted-by":"crossref","first-page":"296","DOI":"10.1186\/1471-2105-10-296","article-title":"NN-align. An artificial neural network-based alignment algorithm for MHC class II peptide binding prediction","volume":"10","author":"Nielsen","year":"2009","journal-title":"BMC Bioinformatics"},{"key":"2024021913362065400_btaa479-B27","doi-asserted-by":"crossref","first-page":"1007","DOI":"10.1110\/ps.0239403","article-title":"Reliable prediction of T-cell epitopes using neural networks with novel sequence representations","volume":"12","author":"Nielsen","year":"2003","journal-title":"Protein Sci"},{"key":"2024021913362065400_btaa479-B28","doi-asserted-by":"crossref","first-page":"129","DOI":"10.1016\/j.cels.2018.05.014","article-title":"MHCflurry: open-source class I MHC binding affinity prediction","volume":"7","author":"O\u2019Donnell","year":"2018","journal-title":"Cell Syst"},{"key":"2024021913362065400_btaa479-B29","doi-asserted-by":"crossref","first-page":"217","DOI":"10.1038\/nature22991","article-title":"An immunogenic personal neoantigen vaccine for patients with melanoma","volume":"547","author":"Ott","year":"2017","journal-title":"Nature"},{"key":"2024021913362065400_btaa479-B30","author":"Peters","year":"2018"},{"key":"2024021913362065400_btaa479-B31","doi-asserted-by":"crossref","first-page":"270","DOI":"10.1186\/s12859-019-2892-4","article-title":"MHCSeqNet: a deep neural network model for universal MHC binding prediction","volume":"20","author":"Phloyphisut","year":"2019","journal-title":"BMC Bioinformatics"},{"key":"2024021913362065400_btaa479-B32","doi-asserted-by":"crossref","first-page":"416","DOI":"10.1016\/j.cell.2018.08.048","article-title":"Evolutionary pressure against MHC class II binding cancer mutations","volume":"175","author":"Pyke","year":"2018","journal-title":"Cell"},{"key":"2024021913362065400_btaa479-B33","doi-asserted-by":"crossref","first-page":"459","DOI":"10.1186\/1471-2105-8-459","article-title":"Predicting peptides binding to MHC class II molecules using multi-objective evolutionary algorithms","volume":"8","author":"Rajapakse","year":"2007","journal-title":"BMC Bioinformatics"},{"key":"2024021913362065400_btaa479-B34","doi-asserted-by":"crossref","first-page":"573","DOI":"10.1016\/j.it.2006.10.005","article-title":"Targeting dendritic cells with biomaterials: developing the next generation of vaccines","volume":"27","author":"Reddy","year":"2006","journal-title":"Trends Immunol"},{"key":"2024021913362065400_btaa479-B35","first-page":"532","volume-title":"Cross-Validation","author":"Refaeilzadeh","year":"2009"},{"key":"2024021913362065400_btaa479-B36","doi-asserted-by":"crossref","first-page":"222","DOI":"10.1038\/nature23003","article-title":"Personalized RNA mutanome vaccines mobilize poly-specific therapeutic immunity against cancer","volume":"547","author":"Sahin","year":"2017","journal-title":"Nature"},{"key":"2024021913362065400_btaa479-B37","doi-asserted-by":"crossref","first-page":"e0118432","DOI":"10.1371\/journal.pone.0118432","article-title":"The precision-recall plot is more informative than the ROC plot when evaluating binary classifiers on imbalanced datasets","volume":"10","author":"Saito","year":"2015","journal-title":"PLoS One"},{"key":"2024021913362065400_btaa479-B38","doi-asserted-by":"crossref","first-page":"2673","DOI":"10.1109\/78.650093","article-title":"Bidirectional recurrent neural networks","volume":"45","author":"Schuster","year":"1997","journal-title":"Trans. Sig. Proc"},{"key":"2024021913362065400_btaa479-B39","doi-asserted-by":"crossref","first-page":"307","DOI":"10.3389\/fpubh.2017.00307","article-title":"Sensitivity, specificity, and predictive values: foundations, pliabilities, and pitfalls in research and practice","volume":"5","author":"Trevethan","year":"2017","journal-title":"Front. Public Health"},{"key":"2024021913362065400_btaa479-B40","first-page":"5998","volume-title":"Advances in Neural Information Processing Systems 30","author":"Vaswani","year":"2017"},{"key":"2024021913362065400_btaa479-B41","doi-asserted-by":"crossref","first-page":"D405","DOI":"10.1093\/nar\/gku938","article-title":"The immune epitope database (IEDB) 3.0","volume":"43","author":"Vita","year":"2015","journal-title":"Nucleic Acids Res"},{"key":"2024021913362065400_btaa479-B42","doi-asserted-by":"crossref","first-page":"e1000048","DOI":"10.1371\/journal.pcbi.1000048","article-title":"A systematic assessment of MHC class II peptide binding predictions and evaluation of a consensus approach","volume":"4","author":"Wang","year":"2008","journal-title":"PLoS Comput. Biol"},{"key":"2024021913362065400_btaa479-B43","first-page":"2048","author":"Xu","year":"2015"},{"key":"2024021913362065400_btaa479-B44","author":"Yang","year":"2018"},{"key":"2024021913362065400_btaa479-B45","doi-asserted-by":"crossref","first-page":"i278","DOI":"10.1093\/bioinformatics\/btz330","article-title":"DeepLigand: accurate prediction of MHC class I ligands using peptide embedding","volume":"35","author":"Zeng","year":"2019","journal-title":"Bioinformatics"},{"key":"2024021913362065400_btaa479-B46","doi-asserted-by":"crossref","first-page":"159","DOI":"10.1016\/j.cels.2019.05.004","article-title":"Quantification of uncertainty in peptide-MHC binding prediction improves high-affinity peptide selection for therapeutic design","volume":"9","author":"Zeng","year":"2019","journal-title":"Cell Syst"},{"key":"2024021913362065400_btaa479-B47","doi-asserted-by":"crossref","first-page":"e1006457","DOI":"10.1371\/journal.pcbi.1006457","article-title":"Systematically benchmarking peptide-MHC binding predictors: from synthetic to naturally processed epitopes","volume":"14","author":"Zhao","year":"2018","journal-title":"PLoS Comput. Biol"},{"key":"2024021913362065400_btaa479-B48","first-page":"207","author":"Zhou","year":"2016"}],"container-title":["Bioinformatics"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/academic.oup.com\/bioinformatics\/article-pdf\/36\/Supplement_1\/i399\/56702814\/bioinformatics_36_supplement1_i399.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"syndication"},{"URL":"https:\/\/academic.oup.com\/bioinformatics\/article-pdf\/36\/Supplement_1\/i399\/56702814\/bioinformatics_36_supplement1_i399.pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2024,2,19]],"date-time":"2024-02-19T13:47:12Z","timestamp":1708350432000},"score":1,"resource":{"primary":{"URL":"https:\/\/academic.oup.com\/bioinformatics\/article\/36\/Supplement_1\/i399\/5870494"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2020,7,1]]},"references-count":48,"journal-issue":{"issue":"Supplement_1","published-print":{"date-parts":[[2020,7,1]]}},"URL":"https:\/\/doi.org\/10.1093\/bioinformatics\/btaa479","relation":{},"ISSN":["1367-4803","1367-4811"],"issn-type":[{"value":"1367-4803","type":"print"},{"value":"1367-4811","type":"electronic"}],"subject":[],"published-other":{"date-parts":[[2020,7]]},"published":{"date-parts":[[2020,7,1]]}}}