{"id":"W4388766745","doi":"10.1016/j.eswa.2023.122500","title":"GGI-DDI: Identification for key molecular substructures by granule learning to interpret predicted drug–drug interactions","year":2023,"lang":"en","type":"article","venue":"Expert Systems with Applications","topic":"Computational Drug Discovery Methods","field":"Computer Science","cited_by":20,"is_retracted":false,"has_abstract":false,"ca_institutions":"University of Regina","funders":"","keywords":"Interpretability; Computer science; Artificial intelligence; Machine learning; Drug-drug interaction; Drug target; Drug; Training set; Key (lock); Pharmacology; Medicine","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":[0.0009298521,0.001150476,0.001347638,0.001882252,0.0003553073,0.001219676,0.001511364,0.0009899777,0.003049602],"category_scores_gemma":[0.002417158,0.000343095,0.001025876,0.001219781,0.0005739845,0.001080615,0.001521919,0.001375459,0.001218864],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0005950349,"about_ca_system_score_gemma":0.001218669,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.001588388,"about_ca_topic_score_gemma":0.002356239,"domain_scores_codex":[0.9997739,0.00003373726,0.00002198862,0.00007402176,0.00006162882,0.00003475728],"domain_scores_gemma":[0.9994038,0.0002244206,0.00009339894,0.0001293158,0.00009057043,0.00005847014],"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":[0.001574824,0.0006929424,0.01974212,0.0008623285,0.0004681485,0.0006259037,0.0002007744,0.09486913,0.07122028,0.01035992,0.02818717,0.7711964],"study_design_scores_gemma":[0.0001430028,0.0002489251,0.003000324,0.00003620617,0.00009098906,0.0001821154,0.00005556012,0.9623157,0.01991087,0.009755161,0.004223604,0.00003758181],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.1662104,0.001222311,0.788528,0.0008335749,0.0002541154,0.0006893425,0.006412067,0.03272488,0.003125274],"genre_scores_gemma":[0.4081527,0.0005477031,0.5788555,0.0003224596,0.00008854835,0.0004401195,0.00843428,0.0007667954,0.002391948],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.003049602,"threshold_uncertainty_score":0.01020199,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.009174085614773241,"score_gpt":0.3018731458323867,"score_spread":0.2926990602176135,"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."}}