{"id":"W4387343780","doi":"10.1016/j.crmeth.2023.100599","title":"RECOVER identifies synergistic drug combinations in vitro through sequential model optimization","year":2023,"lang":"en","type":"article","venue":"Cell Reports Methods","topic":"Computational Drug Discovery Methods","field":"Computer Science","cited_by":18,"is_retracted":false,"has_abstract":true,"ca_institutions":"Hospital for Sick Children; Université de Montréal; Institute for Research in Immunology and Cancer; Mila - Quebec Artificial Intelligence Institute","funders":"Engineering and Physical Sciences Research Council; Bill and Melinda Gates Foundation","keywords":"In silico; Computer science; Benchmarking; Machine learning; Artificial intelligence; Selection (genetic algorithm); Drug; Drug discovery; Computational biology; Bioinformatics; Biology; Pharmacology","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":[0.00106035,0.00215149,0.001607493,0.0008974822,0.0003040461,0.0009408044,0.001114949,0.001419369,0.002600958],"category_scores_gemma":[0.002782944,0.0006716854,0.001650408,0.0006850084,0.0006140894,0.001218059,0.0009665167,0.001451553,0.0007260532],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001339152,"about_ca_system_score_gemma":0.002216377,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.006084125,"about_ca_topic_score_gemma":0.009494458,"domain_scores_codex":[0.9995615,0.0001118628,0.00002749347,0.0001176595,0.0001059136,0.0000755335],"domain_scores_gemma":[0.9989937,0.0006615673,0.00009386036,0.00009336782,0.0001083322,0.0000491942],"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":[0.0001160577,0.00009097758,0.0009687234,0.00008771694,0.00004640923,0.00005491825,0.00001427005,0.9812757,0.00338267,0.001085109,0.0008416796,0.01203585],"study_design_scores_gemma":[0.0000189214,0.00008924536,0.00006836091,0.000004473751,0.00001620122,0.00001445096,0.000007293109,0.9967194,0.001552756,0.001188625,0.0003155462,0.000004763273],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.6597282,0.002414701,0.3145067,0.001554332,0.0001234332,0.0003788931,0.002589402,0.004609889,0.01409446],"genre_scores_gemma":[0.8944145,0.0005232302,0.09728351,0.0005327847,0.00003427178,0.0004302304,0.002888822,0.0002231676,0.003669533],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.006084125,"threshold_uncertainty_score":0.01209742,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.04781760163582092,"score_gpt":0.3688485393359173,"score_spread":0.3210309377000964,"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."}}