{"id":"W4407140906","doi":"10.1186/s13321-025-00962-0","title":"Improving drug repositioning with negative data labeling using large language models","year":2025,"lang":"en","type":"article","venue":"Journal of Cheminformatics","topic":"Computational Drug Discovery Methods","field":"Computer Science","cited_by":8,"is_retracted":false,"has_abstract":true,"ca_institutions":"Université Laval","funders":"","keywords":"Computer science; Drug repositioning; Drug; Data mining; Natural language processing; Data science; Medicine; Pharmacology","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":[0.003608568,0.001520801,0.001038886,0.001403055,0.0005039868,0.00120575,0.001174429,0.0009597035,0.001097362],"category_scores_gemma":[0.008206487,0.0003816141,0.001967724,0.0007350382,0.0005263589,0.001606118,0.0009985799,0.001508416,0.0007356829],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001017014,"about_ca_system_score_gemma":0.002027144,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00530751,"about_ca_topic_score_gemma":0.007990628,"domain_scores_codex":[0.997972,0.0009039337,0.0001654996,0.000466343,0.0003744902,0.0001177672],"domain_scores_gemma":[0.9945885,0.003145659,0.0006152992,0.0006178606,0.0009001631,0.0001325151],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.001137942,0.001401282,0.02477352,0.0005875696,0.0004709648,0.0006106006,0.000260146,0.4324687,0.03152625,0.00387009,0.0120815,0.4908114],"study_design_scores_gemma":[0.00002999889,0.0001616769,0.000937074,0.00002048021,0.00007186148,0.00006709799,0.0000289455,0.9861611,0.008125213,0.002805357,0.001562568,0.00002860552],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.2552929,0.001680137,0.7263515,0.001677223,0.0002800357,0.0004021764,0.0026894,0.007242258,0.004384467],"genre_scores_gemma":[0.7375065,0.0004246054,0.2527961,0.0007742763,0.0001549303,0.0003789353,0.00544268,0.00023593,0.002285996],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.00530751,"threshold_uncertainty_score":0.01908416,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02494095244085814,"score_gpt":0.320357368948671,"score_spread":0.2954164165078128,"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."}}