{"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":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0003229477,0.0001369256,0.0001795964,0.00007229502,0.0000818737,0.00001415843,0.0002063834,0.0002044615,0.000001044433],"category_scores_gemma":[0.0002147949,0.0001340987,0.00006861916,0.0002145608,0.00007322231,0.00000554646,0.00008318094,0.00006104109,0.000001461361],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00001750158,"about_ca_system_score_gemma":0.0001120743,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00001809981,"about_ca_topic_score_gemma":0.000003287846,"domain_scores_codex":[0.9990744,0.00002966825,0.000238256,0.0003398668,0.00008624254,0.0002315631],"domain_scores_gemma":[0.9991997,0.00003351176,0.0001200308,0.0004130121,0.0001803851,0.00005337577],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"bench_or_experimental","study_design_gemma":"bench_or_experimental","study_design_scores_codex":[0.00002973579,0.00006001525,0.001202342,0.0001304458,0.0000415281,6.702373e-7,0.00001531819,0.00001107938,0.9961761,0.001939293,0.0001585202,0.000234902],"study_design_scores_gemma":[0.0005102888,0.0000837229,0.003281491,0.00004098599,0.00002465723,4.178151e-9,0.00003552085,0.002348193,0.9807033,0.00001613496,0.01279058,0.0001651176],"study_design_candidate":"bench_or_experimental","study_design_consensus":"bench_or_experimental","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.7788482,0.002562887,0.2179949,0.00008712986,0.0001046155,0.000239877,0.00008687234,0.00006117832,0.00001433749],"genre_scores_gemma":[0.9786926,0.00006693038,0.02081896,0.0001675438,0.00009863287,0.0001270238,0.000001597408,0.00002016228,0.000006585381],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.1998444,"threshold_uncertainty_score":0.5468385,"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."}}