{"id":"W4415683348","doi":"10.1093/bioinformatics/btaf582","title":"Endowing protein language models with structural knowledge","year":2025,"lang":"en","type":"article","venue":"Bioinformatics","topic":"Biomedical Text Mining and Ontologies","field":"Biochemistry, Genetics and Molecular Biology","cited_by":1,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"Instituto de Ciencias del Mar y Limnología, Universidad Nacional Autónoma de México; Institute for Catastrophic Loss Reduction","keywords":"Code (set theory); Software; Source code; Knowledge representation and reasoning; Natural language; Language model","routes":{"ca_aff":false,"ca_fund":true,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0000755503,0.0001098221,0.000104912,0.00004330137,0.00006951114,0.00002598813,0.000159425,0.0001136878,0.000004542223],"category_scores_gemma":[0.00007197369,0.0000761937,0.00003540226,0.0001013189,0.0000967693,0.000005146902,0.00009509173,0.00007435366,0.000006919238],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.000009736745,"about_ca_system_score_gemma":0.000123535,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.000006151778,"about_ca_topic_score_gemma":0.00001403403,"domain_scores_codex":[0.9994758,0.00001274706,0.00015927,0.00009889079,0.00007039826,0.000182931],"domain_scores_gemma":[0.9996433,0.000007684356,0.00004060408,0.0002187856,0.0000364173,0.00005319645],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"bench_or_experimental","study_design_scores_codex":[0.000464587,0.0001423603,0.001672665,0.00135795,0.0005871395,0.00001887099,0.007678702,0.0004771074,0.1088889,0.01410357,0.02468875,0.8399194],"study_design_scores_gemma":[0.007535197,0.002058163,0.001949349,0.001142891,0.000183244,0.0001158162,0.01736194,0.1355385,0.5677266,0.002647164,0.2616452,0.002096004],"study_design_candidate":"design_other","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.8685247,0.001836715,0.07787958,0.0002558692,0.0001738336,0.0004051248,0.00002266407,0.0001051519,0.05079633],"genre_scores_gemma":[0.9491546,0.000009674503,0.04774486,0.0002432074,0.00005394869,0.00002113287,0.00005291374,0.0000072294,0.002712472],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.8378234,"threshold_uncertainty_score":0.3107089,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01077066642079481,"score_gpt":0.2676589173807811,"score_spread":0.2568882509599862,"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."}}