{"id":"W4406260647","doi":"10.1109/bibm62325.2024.10822459","title":"Synthetic Data from Diffusion Models Improves Drug Discovery Prediction","year":2024,"lang":"en","type":"article","venue":"","topic":"Computational Drug Discovery Methods","field":"Computer Science","cited_by":1,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Waterloo","funders":"","keywords":"Computer science; Drug discovery; Diffusion; Drug; Data mining; Bioinformatics; Pharmacology; Medicine; Physics; Biology","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":[0.002571169,0.0008444295,0.0007539353,0.0008681052,0.0002947095,0.0009064616,0.0008397451,0.001054633,0.001329766],"category_scores_gemma":[0.01069386,0.0004325253,0.0006698323,0.0006007476,0.0006940353,0.001579044,0.0009377023,0.001343905,0.000324757],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001147755,"about_ca_system_score_gemma":0.0009630669,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.004562089,"about_ca_topic_score_gemma":0.004702649,"domain_scores_codex":[0.9993972,0.0002972827,0.00003477939,0.0001246392,0.0001070365,0.00003909092],"domain_scores_gemma":[0.9953408,0.003419152,0.0003430575,0.0004034233,0.0003910355,0.0001025056],"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":[0.0001734574,0.00006839738,0.002453234,0.00008866681,0.00003728413,0.00004441171,0.00003238696,0.9632058,0.001558545,0.005371881,0.001500343,0.02546572],"study_design_scores_gemma":[0.000009500782,0.00002352574,0.0001155431,0.000004528454,0.000003554248,0.000006970809,0.00000302172,0.9959779,0.0006092608,0.002941613,0.0003010347,0.000003485602],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.4102094,0.002298179,0.5722079,0.003220079,0.0002624272,0.0001773149,0.001729088,0.002036765,0.007858822],"genre_scores_gemma":[0.922161,0.000579727,0.07291988,0.0003740233,0.00006395069,0.0001113663,0.002049821,0.00008925265,0.001650923],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.004562089,"threshold_uncertainty_score":0.01359779,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.04330625362961881,"score_gpt":0.2987466478421144,"score_spread":0.2554403942124955,"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."}}