{"id":"W7117357713","doi":"10.1093/jat/bkaf109","title":"Applications of machine learning for the general unknown screening of HRMS data within forensic toxicology","year":2025,"lang":"en","type":"article","venue":"Journal of Analytical Toxicology","topic":"Metabolomics and Mass Spectrometry Studies","field":"Biochemistry, Genetics and Molecular Biology","cited_by":2,"is_retracted":false,"has_abstract":true,"ca_institutions":"Office of the Chief Medical Examiner","funders":"","keywords":"Forensic toxicology; Polysubstance dependence; Forensic science; Drugs of abuse","routes":{"ca_aff":true,"ca_fund":false,"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":[0.0009312411,0.000114597,0.000476141,0.0001507614,0.00009258286,0.000006733112,0.0005567665,0.0001454612,0.00001660239],"category_scores_gemma":[0.0009923925,0.00007712114,0.0001816117,0.0002410952,0.0003050981,0.000004915059,0.0003503077,0.0002230523,1.47665e-7],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00001050871,"about_ca_system_score_gemma":0.0001381551,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.000005624366,"about_ca_topic_score_gemma":0.00006365589,"domain_scores_codex":[0.9987305,0.00009009195,0.0006879334,0.0001979047,0.000104009,0.0001896002],"domain_scores_gemma":[0.9984444,0.0003166493,0.000511153,0.0003655015,0.0003168541,0.00004542286],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"bench_or_experimental","study_design_gemma":"not_applicable","study_design_scores_codex":[0.001701231,0.0006998192,0.01467605,0.0001878003,0.005266261,0.00000497777,0.00006510792,0.006175737,0.7888815,0.1275427,0.01131217,0.04348663],"study_design_scores_gemma":[0.005028406,0.005795013,0.007313738,0.00005971195,0.002438887,0.0001365748,0.0005160252,0.1137568,0.2214908,0.007749826,0.635278,0.000436191],"study_design_candidate":"bench_or_experimental","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.2966121,0.00853496,0.6880825,0.004464947,0.0004941252,0.0005388223,0.0000836846,0.000004438045,0.001184502],"genre_scores_gemma":[0.9808588,0.0008142644,0.01654864,0.0003104686,0.0002883397,0.00001082417,0.00003407544,0.00001081372,0.001123753],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.6842467,"threshold_uncertainty_score":0.3144909,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02961206565556422,"score_gpt":0.3343533993102317,"score_spread":0.3047413336546674,"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."}}