{"id":"W4382930398","doi":"10.1021/acs.chemrestox.3c00062","title":"High Throughput Read-Across for Screening a Large Inventory of Related Structures by Balancing Artificial Intelligence/Machine Learning and Human Knowledge","year":2023,"lang":"en","type":"article","venue":"Chemical Research in Toxicology","topic":"Computational Drug Discovery Methods","field":"Computer Science","cited_by":14,"is_retracted":false,"has_abstract":true,"ca_institutions":"Health Canada","funders":"","keywords":"ADME; Similarity (geometry); In silico; Computer science; Applicability domain; Machine learning; Artificial intelligence; Computational biology; Quantitative structure–activity relationship; Chemistry; Bioinformatics; 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":[0.0057612,0.001753299,0.003467582,0.002899879,0.0006404539,0.002416514,0.002307372,0.001193027,0.00854945],"category_scores_gemma":[0.01265614,0.0009049403,0.002243435,0.002681009,0.000737631,0.002349662,0.002851466,0.00170394,0.004728374],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0009260966,"about_ca_system_score_gemma":0.001224766,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.001050248,"about_ca_topic_score_gemma":0.001695543,"domain_scores_codex":[0.9952155,0.001333101,0.0003017901,0.001276072,0.001658382,0.0002150943],"domain_scores_gemma":[0.988674,0.005426692,0.0008952395,0.00338687,0.00127471,0.0003424333],"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":[0.003182671,0.001217265,0.01031741,0.001978842,0.001006706,0.001146407,0.0003391499,0.0658723,0.2845727,0.01195355,0.01581124,0.6026018],"study_design_scores_gemma":[0.000365262,0.002443865,0.01226847,0.0001599842,0.0006761682,0.001401496,0.0003564121,0.5524502,0.3303362,0.05644263,0.04283638,0.0002628958],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.08315427,0.002741094,0.8736771,0.0005579076,0.0001131629,0.001113517,0.008182551,0.02023778,0.01022258],"genre_scores_gemma":[0.3371803,0.001452907,0.6331643,0.0007978632,0.0001014706,0.001572287,0.0190364,0.001459185,0.005235225],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.00854945,"threshold_uncertainty_score":0.03046852,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1474413194508514,"score_gpt":0.4878320871150905,"score_spread":0.3403907676642391,"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."}}