{"id":"W4385925156","doi":"10.1101/2023.08.15.553429","title":"Benchmarking of Small Molecule Feature Representations for hERG, Nav1.5, and Cav1.2 Cardiotoxicity Prediction","year":2023,"lang":"en","type":"preprint","venue":"bioRxiv (Cold Spring Harbor Laboratory)","topic":"Computational Drug Discovery Methods","field":"Computer Science","cited_by":1,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Alberta","funders":"","keywords":"Cardiotoxicity; hERG; Computer science; Feature (linguistics); Drug discovery; Machine learning; Artificial intelligence; Medicine; Pharmacology; Potassium channel; Bioinformatics; Internal medicine; Biology","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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.001232649,0.001142332,0.0005987752,0.001106424,0.0001902467,0.0005169932,0.001122576,0.0009915252,0.001918218],"category_scores_gemma":[0.002433456,0.0001817436,0.0008666847,0.0008685407,0.0003682581,0.0006248528,0.0005582984,0.0008135647,0.000567157],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0007833637,"about_ca_system_score_gemma":0.0009723133,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.006165207,"about_ca_topic_score_gemma":0.00529773,"domain_scores_codex":[0.9995435,0.0001204235,0.0000341221,0.0001225458,0.0001241324,0.00005531068],"domain_scores_gemma":[0.99918,0.0004122016,0.0000826282,0.0001190703,0.0001467555,0.00005928888],"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":[0.0009913673,0.0006065749,0.008394958,0.000524272,0.0003320601,0.0001964032,0.00002023163,0.8205532,0.01268312,0.001272348,0.01441865,0.1400067],"study_design_scores_gemma":[0.00006100889,0.0004743784,0.001390759,0.00001671189,0.00002960733,0.00004455251,0.000009851792,0.986626,0.009013001,0.0006388725,0.001681467,0.00001370499],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.8285264,0.006414052,0.1201933,0.001110155,0.0003762797,0.0004844146,0.01913872,0.01676678,0.006989861],"genre_scores_gemma":[0.9166411,0.001027316,0.05250936,0.0002513193,0.00004252836,0.0002407462,0.02722587,0.0002178751,0.001843974],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.006165207,"threshold_uncertainty_score":0.01225865,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.03006189695513904,"score_gpt":0.2700285376732328,"score_spread":0.2399666407180937,"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."}}