{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,7,27]],"date-time":"2026-07-27T14:19:39Z","timestamp":1785161979365,"version":"3.55.0"},"reference-count":69,"publisher":"Public Library of Science (PLoS)","issue":"4","license":[{"start":{"date-parts":[[2026,4,30]],"date-time":"2026-04-30T00:00:00Z","timestamp":1777507200000},"content-version":"vor","delay-in-days":0,"URL":"http:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"DOI":"10.13039\/501100006041","name":"Innovate UK","doi-asserted-by":"publisher","award":["INOV01-18"],"award-info":[{"award-number":["INOV01-18"]}],"id":[{"id":"10.13039\/501100006041","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/501100000266","name":"Engineering and Physical Sciences Research Council","doi-asserted-by":"publisher","id":[{"id":"10.13039\/501100000266","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":["www.ploscompbiol.org"],"crossmark-restriction":false},"short-container-title":["PLoS Comput Biol"],"abstract":"<jats:p>Insights gained through interpretation of models trained on the T-cell receptor (TCR) repertoire contribute to advances in understanding of immune-mediated disease. This has the potential to improve diagnostic tests and treatments, particularly for autoimmune diseases. However, TCR repertoire datasets with samples from donors of known autoimmune disease status generally include orders of magnitude fewer samples than TCR sequences. Promising TCR repertoire classification approaches consider relationships between non-identical TCR sequences. In particular, kmer methods demonstrate strong and stable performance for small datasets. We propose a TCR repertoire representation that considers the relationships between amino acids within kmers flexibly and efficiently. XGBoost and logistic regression models are trained and tested on kmer representations of TCR repertoire datasets including samples from patients with coeliac disease as well as donors with previous cytomegalovirus infection. XGBoost models outperform logistic regression, indicating that interactions may be crucial for discriminative ability. We find that a reduced alphabet based on BLOSUM62 can lead to a model with slightly stronger XGBoost testing performance than other kmer features. Though it remains unclear whether there is an amino acid encoding that can substantially improve TCR repertoire classification with reduced alphabet kmers, evidence that this representation enables faster training of XGBoost models in comparison to kmer clusters suggests that our reduced alphabet approach permits wider exploration of amino acid similarity in practice. Finally, we detail motifs which are important in each top-performing XGBoost model and compare them to TCR sequences previously associated with each immune status. We highlight the challenge of interpreting non-linear TCR repertoire classification models trained on kmers which, if overcome, could lead to biomarker discovery for autoimmune diseases.<\/jats:p>","DOI":"10.1371\/journal.pcbi.1014211","type":"journal-article","created":{"date-parts":[[2026,4,30]],"date-time":"2026-04-30T17:48:56Z","timestamp":1777571336000},"page":"e1014211","update-policy":"https:\/\/doi.org\/10.1371\/journal.pcbi.corrections_policy","source":"Crossref","is-referenced-by-count":1,"title":["Evaluating the utility of amino acid similarity-aware kmers to represent TCR repertoires for classification"],"prefix":"10.1371","volume":"22","author":[{"ORCID":"https:\/\/orcid.org\/0000-0002-1196-1195","authenticated-orcid":true,"given":"Hannah","family":"Kockelbergh","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-3242-6017","authenticated-orcid":true,"given":"Shelley C.","family":"Evans","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-3026-4723","authenticated-orcid":true,"given":"Liam","family":"Brierley","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Peter L.","family":"Green","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Andrea L.","family":"Jorgensen","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Elizabeth J.","family":"Soilleux","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-1793-0047","authenticated-orcid":true,"given":"Anna","family":"Fowler","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"340","published-online":{"date-parts":[[2026,4,30]]},"reference":[{"issue":"36","key":"pcbi.1014211.ref001","doi-asserted-by":"crossref","first-page":"13139","DOI":"10.1073\/pnas.1409155111","article-title":"Diversity and clonal selection in the human T-cell repertoire","volume":"111","author":"Q Qi","year":"2014","journal-title":"Proc Natl Acad Sci USA"},{"issue":"10","key":"pcbi.1014211.ref002","doi-asserted-by":"crossref","first-page":"3628","DOI":"10.1073\/pnas.73.10.3628","article-title":"Evidence for somatic rearrangement of immunoglobulin genes coding for variable and constant regions","volume":"73","author":"N Hozumi","year":"1976","journal-title":"Proc Natl Acad Sci U S A"},{"issue":"1","key":"pcbi.1014211.ref003","doi-asserted-by":"crossref","first-page":"523","DOI":"10.1146\/annurev.immunol.16.1.523","article-title":"Ligand recognition by \u03b1\u03b2 T cell receptors","volume":"16","author":"MM Davis","year":"1998","journal-title":"Annu Rev Immunol"},{"issue":"5","key":"pcbi.1014211.ref004","first-page":"659","volume":"49","author":"RO Emerson","journal-title":"Immunosequencing Identifies Signatures of Cytomegalovirus Exposure History and HLA-mediated Effects on the T Cell Repertoire"},{"key":"pcbi.1014211.ref005","author":"TM Snyder"},{"issue":"1","key":"pcbi.1014211.ref006","doi-asserted-by":"crossref","first-page":"14275","DOI":"10.1038\/s41598-021-93608-8","article-title":"Deep learning identifies antigenic determinants of severe SARS-CoV-2 infection within T-cell repertoires","volume":"11","author":"JW Sidhom","year":"2021","journal-title":"Scientific Reports"},{"key":"pcbi.1014211.ref007","doi-asserted-by":"crossref","first-page":"797640","DOI":"10.3389\/fimmu.2022.797640","article-title":"Comparative study of repertoire classification methods reveals data efficiency of k -mer feature extraction","volume":"13","author":"Y Katayama","year":"2022","journal-title":"Front Immunol"},{"key":"pcbi.1014211.ref008","doi-asserted-by":"crossref","unstructured":"Widrich M, Sch\u00e4fl B, Pavlovi\u0107 M, Sandve GK, Hochreiter S, Greiff V, et al. Modern hopfield networks and attention for immune repertoire classification. In: Conference on neural information processing systems. Cold Spring Harbor Laboratory; 2020. https:\/\/doi.org\/10.1101\/2020.04.12.038158","DOI":"10.1101\/2020.04.12.038158"},{"issue":"2","key":"pcbi.1014211.ref009","first-page":"100192","article-title":"Use of machine learning to identify a T cell response to SARS-CoV-2","volume":"2","author":"MS Shoukat","year":"2021","journal-title":"XCRM"},{"issue":"3","key":"pcbi.1014211.ref010","doi-asserted-by":"crossref","DOI":"10.1371\/journal.pone.0265313","article-title":"Dynamic kernel matching for non-conforming data: a case study of T cell receptor datasets","volume":"18","author":"J Ostmeyer","year":"2023","journal-title":"PLOS ONE"},{"key":"pcbi.1014211.ref011","doi-asserted-by":"crossref","DOI":"10.7554\/eLife.68605","article-title":"TCR meta-clonotypes for biomarker discovery with Tcrdist3 enabled identification of public, HLA-restricted clusters of SARS-CoV-2 TCRs","volume":"10","author":"K Mayer-Blackwell","year":"2021","journal-title":"eLife"},{"issue":"6","key":"pcbi.1014211.ref012","doi-asserted-by":"crossref","DOI":"10.1016\/j.immuni.2020.11.004","article-title":"SARS-CoV-2 epitopes are recognized by a public and diverse repertoire of human T cell receptors","volume":"53","author":"AS Shomuradova","year":"2020","journal-title":"Immunity"},{"issue":"8","key":"pcbi.1014211.ref013","doi-asserted-by":"crossref","first-page":"100269","DOI":"10.1016\/j.crmeth.2022.100269","article-title":"Reference-based comparison of adaptive immune receptor repertoires","volume":"2","author":"CR Weber","year":"2022","journal-title":"Cell Reports Methods"},{"issue":"3","key":"pcbi.1014211.ref014","article-title":"Quantitative annotations of T-cell repertoire specificity","volume":"24","author":"J Luo","journal-title":"Bioinformatics"},{"issue":"7661","key":"pcbi.1014211.ref015","doi-asserted-by":"crossref","first-page":"89","DOI":"10.1038\/nature22383","article-title":"Quantifiable predictive features define epitope-specific T cell receptor repertoires","volume":"547","author":"P Dash","year":"2017","journal-title":"Nature"},{"issue":"6","key":"pcbi.1014211.ref016","doi-asserted-by":"crossref","first-page":"1131","DOI":"10.1007\/s10875-021-01045-z","article-title":"Profiling of T cell repertoire in SARS-CoV-2-Infected COVID-19 patients between mild disease and pneumonia","volume":"41","author":"CM Chang","year":"2021","journal-title":"J Clin Immunol"},{"issue":"1","key":"pcbi.1014211.ref017","first-page":"1","article-title":"Machine learning identifies T cell receptor repertoire signatures associated with COVID-19 severity","volume":"6","author":"JJ Park","year":"2023","journal-title":"Commun Biol"},{"key":"pcbi.1014211.ref018","doi-asserted-by":"crossref","DOI":"10.7554\/eLife.63502","article-title":"Longitudinal high-throughput TCR repertoire profiling reveals the dynamics of T-cell memory formation after mild COVID-19 infection","volume":"10","author":"AA Minervina","year":"2021","journal-title":"eLife"},{"issue":"8","key":"pcbi.1014211.ref019","doi-asserted-by":"crossref","DOI":"10.1371\/journal.pone.0249484","article-title":"Impact of HLA type, age and chronic viral infection on peripheral T-cell receptor sharing between unrelated individuals","volume":"16","author":"SA Johnson","year":"2021","journal-title":"PLOS ONE"},{"issue":"7661","key":"pcbi.1014211.ref020","doi-asserted-by":"crossref","first-page":"94","DOI":"10.1038\/nature22976","article-title":"Identifying specificity groups in the T cell receptor repertoire","volume":"547","author":"J Glanville","year":"2017","journal-title":"Nature"},{"key":"pcbi.1014211.ref021","article-title":"Human T cell receptor occurrence patterns encode immune history, genetic background, and receptor specificity","volume":"7","author":"WS DeWitt","year":"2023"},{"issue":"10","key":"pcbi.1014211.ref022","doi-asserted-by":"crossref","DOI":"10.1172\/jci.insight.151849","article-title":"Longitudinal analysis of T cell receptor repertoires reveals shared patterns of antigen-specific response to SARS-CoV-2 infection","volume":"7","author":"RM Gittelman","year":"2022","journal-title":"JCI Insight"},{"key":"pcbi.1014211.ref023","doi-asserted-by":"crossref","unstructured":"Akerman O, Isakov H, Levi R, Psevkin V, Louzoun Y. Counting is almost all you need. Front Immunol. 2023. 13:1031011.\u00a0https:\/\/doi.org\/10.3389\/fimmu.2022.1031011","DOI":"10.3389\/fimmu.2022.1031011"},{"key":"pcbi.1014211.ref024","doi-asserted-by":"crossref","unstructured":"Xu J, Li XX, Yuan N, Li C, Yang JG, Cheng LM. T cell receptor \u03b2 repertoires in patients with COVID-19 reveal disease severity signatures. Front Immunol. 2023;14:1190844. https:\/\/doi.org\/10.3389\/fimmu.2022.1031011","DOI":"10.3389\/fimmu.2023.1190844"},{"key":"pcbi.1014211.ref025","unstructured":"Pradier MF, Prasad N, Chapfuwa P, Ghalebikesabi S, Ilse M, Woodhouse S, et al. AIRIVA: a deep generative model of adaptive immune repertoires. In: Proceedings of machine learning research, 2023. 1\u201336. http:\/\/arxiv.org\/abs\/2304.13737"},{"issue":"6736","key":"pcbi.1014211.ref026","doi-asserted-by":"crossref","DOI":"10.1126\/science.adp2407","article-title":"Disease Diagnostics Using Machine Learning of B Cell and T Cell Receptor Sequences","volume":"387","author":"ME Zaslavsky","year":"2025","journal-title":"Science"},{"key":"pcbi.1014211.ref027","doi-asserted-by":"crossref","first-page":"108621","DOI":"10.1016\/j.clim.2020.108621","article-title":"T cell receptor repertoire as a potential diagnostic marker for celiac disease","volume":"222","author":"Y Yao","year":"2021","journal-title":"Clin Immunol"},{"issue":"3","key":"pcbi.1014211.ref028","doi-asserted-by":"crossref","first-page":"279","DOI":"10.1002\/path.5592","article-title":"Classification of intestinal T-cell receptor repertoires using machine learning methods can identify patients with coeliac disease regardless of dietary gluten status","volume":"253","author":"AD Foers","year":"2023","journal-title":"J Pathology"},{"issue":"12","key":"pcbi.1014211.ref029","doi-asserted-by":"crossref","first-page":"1707","DOI":"10.3390\/biom13121707","article-title":"An interpretable classification model using gluten-specific TCR sequences shows diagnostic potential in coeliac disease","volume":"13","author":"A Fowler","year":"2023","journal-title":"Biomolecules"},{"issue":"7","key":"pcbi.1014211.ref030","first-page":"101125","article-title":"Multimodal immune cell phenotyping in GI biopsies reveals microbiome-related T cell modulations in human GvHD","volume":"4","author":"S Jarosch","year":"2023","journal-title":"Xenobiotic Chemistry Research and Medicine"},{"key":"pcbi.1014211.ref031","article-title":"Characterisation of T cell receptor repertoires in coeliac disease","author":"LW Lee","year":"2022","journal-title":"J Clin Pathol"},{"issue":"8","key":"pcbi.1014211.ref032","doi-asserted-by":"crossref","first-page":"1070","DOI":"10.1136\/annrheumdis-2019-215442","article-title":"T cell receptor \u03b2 repertoires as novel diagnostic markers for systemic lupus erythematosus and rheumatoid arthritis","volume":"78","author":"X Liu","year":"2019","journal-title":"Ann Rheum Dis"},{"issue":"49","key":"pcbi.1014211.ref033","doi-asserted-by":"crossref","DOI":"10.1126\/sciadv.adj6975","article-title":"Tracking DNA-based antigen-specific T cell receptors during progression to type 1 diabetes","volume":"9","author":"AM Mitchell","year":"2023","journal-title":"Sci Adv"},{"issue":"11","key":"pcbi.1014211.ref034","doi-asserted-by":"crossref","first-page":"2194","DOI":"10.1136\/gutjnl-2021-325373","article-title":"A novel unconventional T cell population enriched in Crohn\u2019s disease","volume":"71","author":"E Rosati","year":"2022","journal-title":"Gut"},{"key":"pcbi.1014211.ref035","doi-asserted-by":"crossref","first-page":"109795","DOI":"10.1016\/j.clim.2023.109795","article-title":"Specific T-cell receptor beta-rearrangements of gluten-triggered CD8 T-cells are enriched in celiac disease patients\u2019 duodenal mucosa","author":"V Seitz","year":"2023","journal-title":"Clinical Immunology"},{"issue":"1","key":"pcbi.1014211.ref036","doi-asserted-by":"crossref","first-page":"17977","DOI":"10.1038\/s41598-017-18137-9","article-title":"Deep sequencing of blood and gut T-cell receptor \u03b2-chains reveals gluten-induced immune signatures in celiac disease","volume":"7","author":"DA Yohannes","year":"2017","journal-title":"Scientific Reports"},{"issue":"1","key":"pcbi.1014211.ref037","doi-asserted-by":"crossref","first-page":"159","DOI":"10.1186\/s12859-021-04087-7","article-title":"Clustering based approach for population level identification of condition-associated T-cell receptor \u03b2-Chain CDR3 sequences","volume":"22","author":"DA Yohannes","year":"2021","journal-title":"BMC Bioinformatics"},{"key":"pcbi.1014211.ref038","doi-asserted-by":"crossref","first-page":"414","DOI":"10.1016\/j.ebiom.2019.07.032","article-title":"TCR repertoire and CDR3 motif analyses depict the role of \u03b1\u03b2 T cells in ankylosing spondylitis","volume":"47","author":"M Zheng","year":"2019","journal-title":"EBIOM"},{"issue":"10344","key":"pcbi.1014211.ref039","doi-asserted-by":"crossref","first-page":"2413","DOI":"10.1016\/S0140-6736(22)00794-2","article-title":"Coeliac disease","volume":"399","author":"C Catassi","year":"2022","journal-title":"Lancet"},{"issue":"1","key":"pcbi.1014211.ref040","article-title":"CD, or not CD, that is the question: a digital interobserver agreement study in coeliac disease","volume":"11","author":"J Denholm","year":"2023","journal-title":"BMJ Gastroenterology"},{"issue":"5","key":"pcbi.1014211.ref041","doi-asserted-by":"crossref","DOI":"10.1016\/j.cell.2018.12.039","article-title":"Chronic Inflammation Permanently Reshapes Tissue-Resident Immunity in Celiac Disease","volume":"176","author":"T Mayassi","year":"2019","journal-title":"Cell"},{"issue":"1","key":"pcbi.1014211.ref042","doi-asserted-by":"crossref","first-page":"4699","DOI":"10.1038\/s41467-021-25006-7","article-title":"GIANA allows computationally-efficient TCR clustering and multi-disease repertoire classification by isometric transformation","volume":"12","author":"H Zhang","year":"2021","journal-title":"Nature Commun"},{"issue":"7","key":"pcbi.1014211.ref043","doi-asserted-by":"crossref","first-page":"1671","DOI":"10.1158\/0008-5472.CAN-18-2292","article-title":"Biophysicochemical motifs in T-cell receptor sequences distinguish repertoires from tumor-infiltrating lymphocyte and adjacent healthy tissue","volume":"79","author":"J Ostmeyer","year":"2019","journal-title":"Cancer Research"},{"issue":"3","key":"pcbi.1014211.ref044","doi-asserted-by":"crossref","DOI":"10.1371\/journal.pone.0229569","article-title":"Biophysicochemical motifs in T cell receptor sequences as a potential biomarker for high-grade serous ovarian carcinoma","volume":"15","author":"J Ostmeyer","year":"2020","journal-title":"PLOS ONE"},{"key":"pcbi.1014211.ref045","doi-asserted-by":"crossref","DOI":"10.3389\/fimmu.2021.624230","article-title":"T cell receptor repertoires acquired via routine pap testing may help refine cervical cancer and precancer risk estimates","volume":"12","author":"S Christley","year":"2021","journal-title":"Frontiers in Immunology"},{"issue":"22","key":"pcbi.1014211.ref046","doi-asserted-by":"crossref","first-page":"3181","DOI":"10.1093\/bioinformatics\/btu523","article-title":"Tracking global changes induced in the CD4 T-cell receptor repertoire by immunization with a complex antigen using short stretches of CDR3 protein sequence","volume":"30","author":"N Thomas","year":"2014","journal-title":"Bioinformatics"},{"issue":"42","key":"pcbi.1014211.ref047","doi-asserted-by":"crossref","DOI":"10.1073\/pnas.2408696121","article-title":"Limits on inferring T cell specificity from partial information","volume":"121","author":"J Henderson","year":"2023","journal-title":"Proceed National Acad Sci"},{"issue":"18","key":"pcbi.1014211.ref048","doi-asserted-by":"crossref","first-page":"6395","DOI":"10.1073\/pnas.0408677102","article-title":"Solving the protein sequence metric problem","volume":"102","author":"WR Atchley","year":"2005","journal-title":"Proceed Nat Acad Sci"},{"key":"pcbi.1014211.ref049","doi-asserted-by":"crossref","first-page":"627813","DOI":"10.3389\/fimmu.2021.627813","article-title":"Machine learning analysis of na\u00efve B-cell receptor repertoires stratifies celiac disease patients and controls","volume":"12","author":"O Shemesh","year":"2021","journal-title":"Front Immunol"},{"issue":"1","key":"pcbi.1014211.ref050","doi-asserted-by":"crossref","first-page":"401","DOI":"10.1186\/s12859-017-1814-6","article-title":"Statistical classifiers for diagnosing disease from immune repertoires: a case study using multiple sclerosis","volume":"18","author":"J Ostmeyer","year":"2017","journal-title":"BMC Bioinformatics"},{"issue":"2","key":"pcbi.1014211.ref051","first-page":"129","volume":"28","author":"S Lloyd","journal-title":"Least squares quantization in PCM"},{"issue":"22","key":"pcbi.1014211.ref052","doi-asserted-by":"crossref","first-page":"10915","DOI":"10.1073\/pnas.89.22.10915","article-title":"Amino acid substitution matrices from protein blocks","volume":"89","author":"S Henikoff","year":"1992","journal-title":"Proc Natl Acad Sci U S A"},{"issue":"9","key":"pcbi.1014211.ref053","doi-asserted-by":"crossref","first-page":"1017","DOI":"10.1038\/s41592-022-01578-0","article-title":"VDJdb in the pandemic era: a compendium of T cell receptors specific for SARS-CoV-2","volume":"19","author":"M Goncharov","year":"2022","journal-title":"Nat Methods"},{"issue":"32","key":"pcbi.1014211.ref054","doi-asserted-by":"crossref","first-page":"13073","DOI":"10.1073\/pnas.1311861110","article-title":"Dietary gluten triggers concomitant activation of CD4 and CD8 \u03b1\u03b2 T cells and \u03b3\u03b4 T cells in celiac disease","volume":"110","author":"A Han","year":"2013","journal-title":"Proceed Nat Acad Sci"},{"key":"pcbi.1014211.ref055","doi-asserted-by":"crossref","first-page":"639672","DOI":"10.3389\/fimmu.2021.639672","article-title":"Comprehensive analysis of CDR3 sequences in gluten-specific T-cell receptors reveals a dominant R-motif and several new minor motifs","volume":"12","author":"S Dahal-Koirala","year":"2021","journal-title":"Front Immunol"},{"issue":"1","key":"pcbi.1014211.ref056","first-page":"1","article-title":"\u03b3\u03b4 T cells: origin and fate, subsets, diseases and immunotherapy","volume":"8","author":"Y Hu","year":"2023","journal-title":"Nature Rev Immunol"},{"issue":"1","key":"pcbi.1014211.ref057","article-title":"tcrBLOSUM: an amino acid substitution matrix for sensitive alignment of distant epitope-specific TCRs","volume":"26","author":"A Postovskaya","year":"2023","journal-title":"Bioinformatics"},{"key":"pcbi.1014211.ref058","author":"A Abu-Rmileh"},{"issue":"1","key":"pcbi.1014211.ref059","doi-asserted-by":"crossref","first-page":"323","DOI":"10.1084\/jem.179.1.323","article-title":"CDR3 Length in Antigen-Specific Immune Receptors","volume":"179","author":"EP Rock","year":"1999","journal-title":"J Exp Med"},{"issue":"5","key":"pcbi.1014211.ref060","doi-asserted-by":"crossref","first-page":"380","DOI":"10.1038\/nmeth.3364","article-title":"MiXCR: software for comprehensive adaptive immunity profiling","volume":"12","author":"DA Bolotin","year":"2015","journal-title":"Nat Methods"},{"issue":"8","key":"pcbi.1014211.ref061","doi-asserted-by":"crossref","first-page":"1035","DOI":"10.1038\/nbt0804-1035","article-title":"Where did the BLOSUM62 alignment score matrix come from?","volume":"22","author":"SR Eddy","year":"2004","journal-title":"Nat Biotechnol"},{"issue":"10","key":"pcbi.1014211.ref062","doi-asserted-by":"crossref","first-page":"1194","DOI":"10.1038\/s41587-020-0505-4","article-title":"Analyzing the Mycobacterium tuberculosis immune response by T-cell receptor clustering with GLIPH2 and genome-wide antigen screening","volume":"38","author":"H Huang","year":"2020","journal-title":"Nature Biotechnology"},{"issue":"6","key":"pcbi.1014211.ref063","first-page":"1359","article-title":"Investigation of antigen-specific T-cell receptor clusters in human cancers","volume":"26","author":"H Zhang","year":"2019","journal-title":"Cancer Research"},{"issue":"18","key":"pcbi.1014211.ref064","doi-asserted-by":"crossref","first-page":"1980","DOI":"10.1093\/bioinformatics\/btn382","article-title":"Optimizing amino acid groupings for GPCR classification","volume":"24","author":"MN Davies","year":"2008","journal-title":"Bioinformatics"},{"issue":"3","key":"pcbi.1014211.ref065","doi-asserted-by":"crossref","first-page":"261","DOI":"10.1038\/s41592-019-0686-2","article-title":"SciPy 1.0: fundamental algorithms for scientific computing in python","volume":"17","author":"P Virtanen","year":"2020","journal-title":"Nature Methods"},{"key":"pcbi.1014211.ref066","first-page":"2825","article-title":"Scikit-learn: machine learning in Python","volume":"12","author":"F Pedregosa","year":"2011","journal-title":"J Mach Learn Res"},{"key":"pcbi.1014211.ref067","doi-asserted-by":"crossref","unstructured":"Chen T, Guestrin C. Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, 2016. 785\u201394. http:\/\/arxiv.org\/abs\/1603.02754","DOI":"10.1145\/2939672.2939785"},{"issue":"3","key":"pcbi.1014211.ref068","doi-asserted-by":"crossref","first-page":"90","DOI":"10.1109\/MCSE.2007.55","article-title":"Matplotlib: a 2D graphics environment","volume":"9","author":"JD Hunter","year":"2007","journal-title":"Comput Sci Eng"},{"key":"pcbi.1014211.ref069","doi-asserted-by":"crossref","unstructured":"Akiba T, Sano S, Yanase T, Ohta T, Koyama M. Optuna: a next-generation hyperparameter optimisation framework. In: Proceedings of the 25th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. 2623\u201331. http:\/\/arxiv.org\/abs\/1907.10902","DOI":"10.1145\/3292500.3330701"}],"container-title":["PLOS Computational Biology"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/dx.plos.org\/10.1371\/journal.pcbi.1014211","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,4,30]],"date-time":"2026-04-30T17:49:10Z","timestamp":1777571350000},"score":1,"resource":{"primary":{"URL":"https:\/\/dx.plos.org\/10.1371\/journal.pcbi.1014211"}},"subtitle":[],"editor":[{"given":"Shanfeng","family":"Zhu","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"editor"}]}],"short-title":[],"issued":{"date-parts":[[2026,4,30]]},"references-count":69,"journal-issue":{"issue":"4","published-online":{"date-parts":[[2026,4,30]]}},"URL":"https:\/\/doi.org\/10.1371\/journal.pcbi.1014211","relation":{},"ISSN":["1553-7358"],"issn-type":[{"value":"1553-7358","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,4,30]]}}}