{"id":"W7116973535","doi":"10.64898/2025.12.20.695723","title":"Application of Large Language Models for Annotating Genes into Reactome Pathways","year":2025,"lang":"en","type":"article","venue":"bioRxiv (Cold Spring Harbor Laboratory)","topic":"Biomedical Text Mining and Ontologies","field":"Biochemistry, Genetics and Molecular Biology","cited_by":1,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Toronto; Ontario Institute for Cancer Research","funders":"National Institutes of Health","keywords":"Workflow; Data curation; Similarity (geometry); Resource (disambiguation); Semantic similarity; Semantics (computer science)","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":[0.005021443,0.001785476,0.000768155,0.003090182,0.001231315,0.002914255,0.002032382,0.000744524,0.002856423],"category_scores_gemma":[0.01254675,0.0008780826,0.004551888,0.002586923,0.0007019262,0.002313572,0.002561543,0.001715477,0.001537198],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.002636172,"about_ca_system_score_gemma":0.003467923,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.006956765,"about_ca_topic_score_gemma":0.008508965,"domain_scores_codex":[0.9969993,0.00106051,0.0004895603,0.0007405511,0.0006083509,0.0001017605],"domain_scores_gemma":[0.9935321,0.003605467,0.0005603846,0.001146808,0.001025228,0.000130113],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"simulation_or_modeling","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.001193814,0.000522762,0.01420436,0.003362008,0.0009277805,0.002716843,0.002592986,0.4859879,0.1029599,0.1276219,0.02180754,0.2361023],"study_design_scores_gemma":[0.000073241,0.00007897368,0.0008917714,0.0001291465,0.0001699679,0.0001954515,0.0002121142,0.8635103,0.04392397,0.05306229,0.03764954,0.0001031737],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.01233629,0.0001892683,0.9626456,0.0003566091,0.0001058201,0.0004566972,0.004946367,0.01718041,0.0017828],"genre_scores_gemma":[0.07601587,0.0003538713,0.9097632,0.0002123227,0.00003682449,0.0008379216,0.009956935,0.001485058,0.001338082],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.006956765,"threshold_uncertainty_score":0.02655619,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.009973064382286952,"score_gpt":0.2524537599215689,"score_spread":0.2424806955392819,"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."}}