{"id":"W3096410258","doi":"10.1016/j.comtox.2020.100142","title":"Evaluation of quantitative structure property relationship algorithms for predicting plasma protein binding in humans","year":2020,"lang":"en","type":"article","venue":"Computational Toxicology","topic":"Computational Drug Discovery Methods","field":"Computer Science","cited_by":28,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Waterloo","funders":"Natural Sciences and Engineering Research Council of Canada; Canadian Network for Research and Innovation in Machining Technology, Natural Sciences and Engineering Research Council of Canada; U.S. Environmental Protection Agency","keywords":"Quantitative structure–activity relationship; Lipophilicity; Chemistry; Molecular descriptor; Polar surface area; Training set; Stereochemistry; Molecule; Organic chemistry; Artificial intelligence; Computer science","routes":{"ca_aff":true,"ca_fund":true,"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":[0.001871451,0.000157746,0.0002712179,0.0002532084,0.0001326834,0.00004543876,0.0004446965,0.0001089341,0.00001583771],"category_scores_gemma":[0.003372902,0.0001426496,0.00006923895,0.0007517735,0.00008237595,0.0004805286,0.0001502785,0.0002106703,0.000004490006],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001969566,"about_ca_system_score_gemma":0.0007682048,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.000005046646,"about_ca_topic_score_gemma":0.00001832481,"domain_scores_codex":[0.9970079,0.0009024724,0.0005940832,0.000521835,0.0007508269,0.000222898],"domain_scores_gemma":[0.9966932,0.001957272,0.0003300454,0.0001301689,0.0008181962,0.00007110991],"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":[0.00003717002,0.00005899419,0.002141174,0.00004600294,0.00003039184,0.000001198607,0.001962985,0.8698528,0.001649562,0.1158555,0.00003558001,0.008328581],"study_design_scores_gemma":[0.00118484,0.0003079448,0.0225606,0.00003156279,0.00001765434,0.000002862516,0.00008751901,0.8553522,0.0007326784,0.1195639,0.00002600825,0.0001322496],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.4577729,0.00002469859,0.5392889,0.001598263,0.0001292115,0.00100412,0.00003975201,0.00003948573,0.0001026407],"genre_scores_gemma":[0.641572,9.566666e-8,0.3581313,0.00007986855,0.00004444481,0.0001072279,0.00004984346,0.000009101494,0.000006147221],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.1837991,"threshold_uncertainty_score":0.5817083,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.2105543463801471,"score_gpt":0.4011006702650156,"score_spread":0.1905463238848686,"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."}}