{"id":"W4321499530","doi":"10.2533/chimia.2023.22","title":"Putting Chemical Knowledge to Work in Machine Learning for Reactivity","year":2023,"lang":"en","type":"article","venue":"CHIMIA International Journal for Chemistry","topic":"Computational Drug Discovery Methods","field":"Computer Science","cited_by":13,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"Vetenskapsrådet; University of Toronto","keywords":"Cheminformatics; Computer science; Artificial neural network; Machine learning; Artificial intelligence; Chemometrics; Deep learning; Chemistry; Computational chemistry","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":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.001139104,0.0001485438,0.000172472,0.0001696814,0.0001153837,0.0002516661,0.001077976,0.00006973361,0.000008499204],"category_scores_gemma":[0.001904832,0.0001566537,0.0001813834,0.0004834411,0.00001589289,0.0002938628,0.0003775502,0.0003705065,0.00001540007],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0003202324,"about_ca_system_score_gemma":0.0001535384,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.000002003973,"about_ca_topic_score_gemma":0.000001017941,"domain_scores_codex":[0.9986266,0.00003542859,0.0003550915,0.0003529871,0.0003187119,0.0003111259],"domain_scores_gemma":[0.998137,0.001154279,0.000146828,0.0001284625,0.0002937337,0.0001396779],"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":[0.0006982328,0.00047289,0.006207167,0.0001997593,0.0002650798,0.00009026564,0.002845464,0.1529165,0.2922567,0.003639998,0.008722248,0.5316857],"study_design_scores_gemma":[0.002156243,0.00004389869,0.004936335,0.0003402279,0.00001000662,0.0002141782,0.0000552129,0.7558379,0.1677178,0.03109555,0.03709328,0.0004993122],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.5379857,0.00004389715,0.4562408,0.003388555,0.00148407,0.0002060427,0.00001479868,0.0001254986,0.0005107046],"genre_scores_gemma":[0.8687612,0.000007943969,0.1286949,0.0001173769,0.001192765,0.0001096734,0.00007412268,0.00003076525,0.001011268],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.6029214,"threshold_uncertainty_score":0.6388153,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.0388675585431783,"score_gpt":0.3707576066492097,"score_spread":0.3318900481060314,"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."}}