{"id":"W4410477113","doi":"10.1101/2025.05.17.654017","title":"SyntheMol-RL: a flexible reinforcement learning framework for designing novel and synthesizable antibiotics","year":2025,"lang":"en","type":"preprint","venue":"bioRxiv (Cold Spring Harbor Laboratory)","topic":"Computational Drug Discovery Methods","field":"Computer Science","cited_by":4,"is_retracted":false,"has_abstract":true,"ca_institutions":"McMaster University","funders":"Natural Sciences and Engineering Research Council of Canada; Canadian Institutes of Health Research; Stanford Bio-X","keywords":"Reinforcement learning; Computer architecture; Computer science; Artificial intelligence","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":[0.0009389973,0.0008455423,0.0007622743,0.0004196719,0.0002664982,0.0006638979,0.001817973,0.001075631,0.004430416],"category_scores_gemma":[0.001666877,0.000568606,0.001001705,0.0002707018,0.0007824908,0.0006574346,0.0008666897,0.001160496,0.0008067856],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0008972325,"about_ca_system_score_gemma":0.001210952,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.002890933,"about_ca_topic_score_gemma":0.004492507,"domain_scores_codex":[0.9997491,0.00008430427,0.00001210171,0.00005345068,0.00007052167,0.00003034783],"domain_scores_gemma":[0.9996021,0.0002340526,0.00004503334,0.00004056088,0.0000416183,0.00003666573],"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":[0.00004245122,0.00003518319,0.0002498292,0.00006020217,0.00001901421,0.00004993649,0.00001413543,0.9727614,0.003086848,0.006293264,0.0005873386,0.01680039],"study_design_scores_gemma":[0.00001142601,0.00001548069,0.000009978598,0.000003057753,0.000002822496,0.000004793961,0.000001318854,0.9972137,0.0004867345,0.001788628,0.0004597075,0.000002393005],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.02227926,0.0003908003,0.9680851,0.0002871025,0.000047751,0.0001040392,0.0002332385,0.00384209,0.004730722],"genre_scores_gemma":[0.498467,0.0004239347,0.4958094,0.0003205935,0.00004185916,0.0005324592,0.0004956793,0.000548429,0.003360631],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.004430416,"threshold_uncertainty_score":0.01482123,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.0283796774691102,"score_gpt":0.2783065753726102,"score_spread":0.2499268979035,"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."}}