{"id":"W4401214408","doi":"10.1186/s12859-024-05861-z","title":"Hybrid fragment-SMILES tokenization for ADMET prediction in drug discovery","year":2024,"lang":"en","type":"article","venue":"BMC Bioinformatics","topic":"Computational Drug Discovery Methods","field":"Computer Science","cited_by":26,"is_retracted":false,"has_abstract":true,"ca_institutions":"National Research Council Canada; Brock University","funders":"National Research Council Canada; Natural Sciences and Engineering Research Council of Canada; Canada Foundation for Innovation","keywords":"Lexical analysis; Computer science; In silico; Drug discovery; Computational biology; Transformer; Artificial intelligence; Machine learning; Bioinformatics; Biology; Genetics","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":[0.0007032029,0.0001482375,0.0001444208,0.0003354285,0.00006829795,0.0006811015,0.0003985708,0.00003667875,0.000005058258],"category_scores_gemma":[0.0001751738,0.0001375774,0.00009196932,0.000482497,0.00002487528,0.003479523,0.0001753722,0.0001043142,0.00004532875],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.000140002,"about_ca_system_score_gemma":0.0002563888,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.000008708838,"about_ca_topic_score_gemma":0.00001052325,"domain_scores_codex":[0.9986577,0.00005685955,0.0005219011,0.0002170999,0.0003154786,0.0002309745],"domain_scores_gemma":[0.9988331,0.0007011216,0.00007894268,0.0002804401,0.00005671559,0.00004973486],"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":[0.00002842056,0.000110862,0.0009028364,0.001463869,0.00004113827,0.000005529098,0.004480536,0.559893,0.0000288187,0.3190725,0.02162464,0.0923479],"study_design_scores_gemma":[0.0002376419,0.00003078918,0.001142447,0.0001225141,0.000007771047,0.00001181734,0.000101971,0.9741427,0.0004150148,0.01857018,0.005071352,0.0001458174],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.01473784,0.000168375,0.9819909,0.0002013662,0.001259444,0.0004739397,0.0001183423,0.0002398646,0.0008099413],"genre_scores_gemma":[0.09328656,0.00002953949,0.9054232,0.0002079387,0.000166615,0.0001016743,0.0002131442,0.00001815178,0.0005531905],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.4142497,"threshold_uncertainty_score":0.6567877,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01916693910531592,"score_gpt":0.2797357114306657,"score_spread":0.2605687723253498,"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."}}