{"id":"W4415222531","doi":"10.1093/pnasnexus/pgaf329","title":"Stereochemistry-aware string-based molecular generation","year":2025,"lang":"en","type":"article","venue":"PNAS Nexus","topic":"Various Chemistry Research Topics","field":"Chemistry","cited_by":1,"is_retracted":false,"has_abstract":true,"ca_institutions":"Canadian Institute for Advanced Research; Vector Institute; University of Toronto","funders":"Alliance de recherche numérique du Canada; Vetenskapsrådet; Ministère de l'Économie, de la Science et de l'Innovation - Québec; Natural Sciences and Engineering Research Council of Canada; Canadian Institute for Advanced Research","keywords":"Generative grammar; Chemical space; Generative model; Reinforcement learning; Drug discovery; Computational model","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":[0.00008810076,0.0002086831,0.0001788694,0.00004208113,0.0001173979,0.0001202844,0.000388927,0.0002071039,0.0009106701],"category_scores_gemma":[0.0002004226,0.0002338672,0.0001034072,0.0001984064,0.00005786119,0.00004802112,0.0001145362,0.0003485043,0.00003032834],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0002120447,"about_ca_system_score_gemma":0.0003254711,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00003087566,"about_ca_topic_score_gemma":0.000008241283,"domain_scores_codex":[0.998587,0.00001578449,0.0002479062,0.0004291311,0.0003326513,0.0003875881],"domain_scores_gemma":[0.9989063,0.0000506118,0.00006127547,0.0007178891,0.0001391237,0.0001247854],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"bench_or_experimental","study_design_gemma":"bench_or_experimental","study_design_scores_codex":[0.00001619563,0.000083634,0.001113114,0.0003809252,0.00005171398,0.00006828635,0.0000166517,0.0006451028,0.9920303,0.0003982831,0.00243512,0.002760698],"study_design_scores_gemma":[0.0006499906,0.000005828807,0.00001503021,0.00006159927,0.00002271987,0.00000351693,0.0000294265,0.01652013,0.9706323,0.0003959242,0.01147086,0.0001926483],"study_design_candidate":"bench_or_experimental","study_design_consensus":"bench_or_experimental","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.7927764,0.0005392698,0.01543775,0.001564671,0.0001620198,0.0001526293,0.00008101288,0.000397474,0.1888888],"genre_scores_gemma":[0.986385,0.000003817744,0.0003158413,0.0002030937,0.0001846823,0.00005323021,0.0002571261,0.00002353709,0.01257361],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.1936087,"threshold_uncertainty_score":0.9971199,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01852587094180864,"score_gpt":0.2887418240272049,"score_spread":0.2702159530853963,"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."}}