{"id":"W7117134055","doi":"10.1038/s41598-025-33443-3","title":"Bidirectional reinforcement learning neural network for constrained molecular design","year":2025,"lang":"en","type":"article","venue":"Scientific Reports","topic":"Computational Drug Discovery Methods","field":"Computer Science","cited_by":1,"is_retracted":false,"has_abstract":true,"ca_institutions":"Defence Research and Development Canada; National Research Council Canada","funders":"National Research Council Canada; Defence Research and Development Canada","keywords":"Reinforcement learning; Constraint (computer-aided design); Process (computing); Chemical space; Artificial neural network; Space (punctuation); Function (biology)","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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.000795896,0.0006005693,0.0006293128,0.0002585299,0.0002118163,0.0003846054,0.000946007,0.0006968181,0.001877007],"category_scores_gemma":[0.001867337,0.0003012165,0.0003378208,0.000250349,0.0005910064,0.0004893976,0.0007487775,0.0008369469,0.0003220052],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0008078492,"about_ca_system_score_gemma":0.0008503579,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.003061708,"about_ca_topic_score_gemma":0.003725183,"domain_scores_codex":[0.9997396,0.00009610302,0.00001111506,0.00004723871,0.00007548509,0.00003057874],"domain_scores_gemma":[0.9995375,0.0002517134,0.00006567602,0.00003985158,0.00007680129,0.0000284792],"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":[0.00002461375,0.00002115368,0.0001483892,0.00003114206,0.00001135168,0.00002567164,0.00001083722,0.9773327,0.001727462,0.005459088,0.0002653728,0.01494223],"study_design_scores_gemma":[0.000003580495,0.00001273806,0.00001139904,0.000001570968,0.000001606874,0.000002950176,6.270814e-7,0.9981887,0.0002657192,0.001371095,0.0001386745,0.000001419598],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.0298506,0.0003445823,0.9651469,0.0001577487,0.00003318996,0.00004750992,0.00004672444,0.000601494,0.003771243],"genre_scores_gemma":[0.8156618,0.0002358114,0.1799385,0.0001630823,0.00002062459,0.0002574665,0.0001354112,0.0001130902,0.003474207],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.003061708,"threshold_uncertainty_score":0.00627923,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02565232698269359,"score_gpt":0.3028818870649685,"score_spread":0.2772295600822749,"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."}}