{"id":"W7117296782","doi":"10.1021/acs.jproteome.5c00506","title":"Protein Language Models: Applications and Perspectives","year":2025,"lang":"en","type":"article","venue":"Journal of Proteome Research","topic":"Machine Learning in Bioinformatics","field":"Biochemistry, Genetics and Molecular Biology","cited_by":5,"is_retracted":false,"has_abstract":true,"ca_institutions":"Université Laval; Centre hospitalier de l'Université Laval","funders":"","keywords":"Key (lock); Sequence (biology); Protein sequencing; Drug discovery; Proteomics; Language model","routes":{"ca_aff":true,"ca_fund":false,"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":[0.004186878,0.00116621,0.001152628,0.002052031,0.0004725683,0.003744452,0.00198866,0.002188787,0.003945034],"category_scores_gemma":[0.01450038,0.0006470997,0.001198811,0.002927436,0.001938545,0.00533192,0.001900572,0.003390007,0.002677419],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001748573,"about_ca_system_score_gemma":0.001290467,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.005154603,"about_ca_topic_score_gemma":0.002418374,"domain_scores_codex":[0.9984101,0.0007829568,0.00006758634,0.0002594513,0.0003873789,0.00009240407],"domain_scores_gemma":[0.991735,0.00590891,0.0002208528,0.0006631852,0.001244663,0.0002274351],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"theoretical_or_conceptual","study_design_gemma":"theoretical_or_conceptual","study_design_scores_codex":[0.0001773692,0.000214017,0.003689655,0.0008714951,0.0002256215,0.0004524904,0.0003038639,0.1046671,0.001035262,0.4888841,0.04539251,0.3540865],"study_design_scores_gemma":[0.00001800975,0.00004975167,0.0005845916,0.0002817697,0.00002636864,0.0002942054,0.0001245373,0.3080639,0.0005273089,0.6403918,0.04958389,0.00005374812],"study_design_candidate":"theoretical_or_conceptual","study_design_consensus":"theoretical_or_conceptual","genre_codex":"methods","genre_gemma":"review","genre_scores_codex":[0.01543941,0.2571643,0.6279832,0.05343409,0.001661153,0.0001053418,0.001589661,0.002381712,0.04024113],"genre_scores_gemma":[0.4519982,0.2448622,0.2613739,0.00691935,0.007145377,0.000384939,0.003612705,0.0007520714,0.02295127],"genre_candidate":"review","genre_consensus":null,"teacher_disagreement_score":0.005154603,"threshold_uncertainty_score":0.02214265,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.031509319623001,"score_gpt":0.3799376186048076,"score_spread":0.3484282989818066,"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."}}