{"id":"W4408690205","doi":"10.1039/d5sc01476j","title":"Directly optimizing for synthesizability in generative molecular design using retrosynthesis models","year":2025,"lang":"en","type":"article","venue":"Chemical Science","topic":"Computational Drug Discovery Methods","field":"Computer Science","cited_by":22,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"Natural Sciences and Engineering Research Council of Canada; Schweizerischer Nationalfonds zur Förderung der Wissenschaftlichen Forschung","keywords":"Retrosynthetic analysis; Generative grammar; Computer science; Artificial intelligence; Machine learning; Engineering; Chemistry; Stereochemistry","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":[0.001738378,0.0011383,0.0009971657,0.000742608,0.0004860158,0.001412774,0.001270632,0.0009866758,0.007727019],"category_scores_gemma":[0.004916376,0.0007119101,0.001261132,0.0005422643,0.00098626,0.001494902,0.001089907,0.001642081,0.001367901],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001271776,"about_ca_system_score_gemma":0.001943502,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.001884912,"about_ca_topic_score_gemma":0.005508773,"domain_scores_codex":[0.9994672,0.0001792365,0.00002395116,0.000122498,0.0001525551,0.00005457133],"domain_scores_gemma":[0.9979427,0.001573004,0.0001008413,0.0002208273,0.0001075912,0.00005500681],"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":[0.00009773181,0.00006955164,0.001182333,0.0002599031,0.00005450296,0.0001079048,0.00006414453,0.8895657,0.00604419,0.05635444,0.00174039,0.04445932],"study_design_scores_gemma":[0.00002081351,0.00004353416,0.00007413,0.00002143508,0.00002800598,0.00003138485,0.0000138537,0.9611213,0.002637545,0.03354165,0.00245662,0.000009785938],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.02500519,0.0005161005,0.9628315,0.0003891383,0.00005985235,0.0001458769,0.0004551163,0.001566067,0.009031164],"genre_scores_gemma":[0.4803571,0.0006700161,0.5104154,0.0004675758,0.00004553442,0.0005666729,0.001114256,0.0009324066,0.005431001],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.007727019,"threshold_uncertainty_score":0.02584946,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.06296741142680679,"score_gpt":0.3462113161996178,"score_spread":0.283243904772811,"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."}}