{"id":"W4223937330","doi":"10.3390/molecules27082513","title":"A Consensus Compound/Bioactivity Dataset for Data-Driven Drug Design and Chemogenomics","year":2022,"lang":"en","type":"article","venue":"Molecules","topic":"Computational Drug Discovery Methods","field":"Computer Science","cited_by":29,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"European Commission; European Federation of Pharmaceutical Industries and Associations; Diamond Light Source; Innovative Medicines Initiative; McGill University","keywords":"chEMBL; PubChem; Drug discovery; Computer science; Chemical space; Computational biology; Combinatorial chemistry; Chemistry; Bioinformatics; Biology","routes":{"ca_aff":false,"ca_fund":true,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.00409417,0.001630247,0.001922842,0.00455813,0.0007999467,0.001728443,0.003431785,0.002483391,0.005496175],"category_scores_gemma":[0.01173435,0.0006424669,0.001875491,0.006266035,0.0005175506,0.001517831,0.002039445,0.002518497,0.004125047],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001315061,"about_ca_system_score_gemma":0.005405298,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.004338335,"about_ca_topic_score_gemma":0.008260465,"domain_scores_codex":[0.9970856,0.0005301944,0.0004648434,0.0006926064,0.001053173,0.0001736063],"domain_scores_gemma":[0.9935744,0.002227828,0.0006995727,0.001516198,0.001445539,0.0005365099],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"not_applicable","study_design_gemma":"not_applicable","study_design_scores_codex":[0.00324457,0.002809445,0.04996702,0.01545866,0.00208502,0.001422417,0.0002875721,0.1032217,0.05918698,0.01641673,0.522364,0.2235359],"study_design_scores_gemma":[0.001728436,0.001387701,0.03629713,0.001032869,0.001181645,0.001568693,0.0002721222,0.1722756,0.06994247,0.01977271,0.6941777,0.0003629531],"study_design_candidate":"not_applicable","study_design_consensus":"not_applicable","genre_codex":"dataset","genre_gemma":"dataset","genre_scores_codex":[0.03151204,0.002374833,0.0506622,0.0005832273,0.0001936019,0.000979971,0.90185,0.006876096,0.004967969],"genre_scores_gemma":[0.02575965,0.0004967115,0.04330752,0.0002077333,0.00002252945,0.0008238083,0.92877,0.0002457019,0.0003663404],"genre_candidate":"dataset","genre_consensus":"dataset","teacher_disagreement_score":0.005496175,"threshold_uncertainty_score":0.02165228,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.07759341046365603,"score_gpt":0.321248517982168,"score_spread":0.2436551075185119,"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."}}