{"id":"W4310009559","doi":"10.1101/2022.11.23.517563","title":"Machine learning analysis of the T cell receptor repertoire identifies sequence features that predict self-reactivity","year":2022,"lang":"en","type":"preprint","venue":"bioRxiv (Cold Spring Harbor Laboratory)","topic":"T-cell and B-cell Immunology","field":"Immunology and Microbiology","cited_by":3,"is_retracted":false,"has_abstract":true,"ca_institutions":"Université de Montréal; Hôpital Maisonneuve-Rosemont; McGill University","funders":"Natural Sciences and Engineering Research Council of Canada; Canadian Institutes of Health Research; Exzellenzclusters Entzündungsforschung; Universität zu Lübeck; Deutsche Forschungsgemeinschaft; Nederlandse Organisatie voor Wetenschappelijk Onderzoek; McGill University","keywords":"T-cell receptor; Repertoire; Reactivity (psychology); Receptor; Biology; Sequence (biology); Major histocompatibility complex; T cell; Computational biology; Immunology; Antigen; Genetics; Immune system; Medicine; Physics","routes":{"ca_aff":true,"ca_fund":true,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":false},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0009612884,0.0004324601,0.0003703301,0.0009781981,0.0001702632,0.000641679,0.0002184334,0.00039102,0.001138804],"category_scores_gemma":[0.002040419,0.0001113553,0.0004520934,0.0004398587,0.0002416819,0.000339731,0.0002063861,0.000640242,0.0005778725],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0002652589,"about_ca_system_score_gemma":0.0002354539,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0004971505,"about_ca_topic_score_gemma":0.0004056937,"domain_scores_codex":[0.9996583,0.0001233933,0.00002045425,0.00009421212,0.00006247211,0.00004116341],"domain_scores_gemma":[0.9986916,0.0008541983,0.0001520707,0.0001164617,0.0001239194,0.00006180412],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"observational","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.001087858,0.000510361,0.276182,0.0002631196,0.0003583962,0.0002624784,0.0001005162,0.2723111,0.2601493,0.00120052,0.002360836,0.1852135],"study_design_scores_gemma":[0.00001764371,0.0001579322,0.05837029,0.00001429584,0.0000341138,0.000132345,0.00003234108,0.915367,0.02256834,0.002711339,0.0005811225,0.0000133507],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9643178,0.0002526272,0.03298199,0.0001432685,0.00001820139,0.00001928071,0.0007409506,0.0003173095,0.001208588],"genre_scores_gemma":[0.989738,0.00005358062,0.008725315,0.00002932347,0.0000131825,0.00001767766,0.001137921,0.0000178035,0.0002671757],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.001138804,"threshold_uncertainty_score":0.005083859,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01174076816555322,"score_gpt":0.2075527813867186,"score_spread":0.1958120132211654,"validation_status":"score_only:v0-immature-baseline","note":"Baseline scores from an immature model (maturity gate not passed). Scores rank; they never assert a category."}}