{"id":"W3089057837","doi":"10.1101/2020.09.25.314138","title":"Improving drug safety predictions by reducing poor analytical practices","year":2020,"lang":"en","type":"preprint","venue":"bioRxiv (Cold Spring Harbor Laboratory)","topic":"Statistical Methods in Clinical Trials","field":"Mathematics","cited_by":2,"is_retracted":false,"has_abstract":true,"ca_institutions":"Prioris.ai (Canada)","funders":"","keywords":"Risk analysis (engineering); Computer science; Reuse; Simple (philosophy); Drug; Medicine; Engineering; Pharmacology","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":["metaresearch","metaepi_narrow","research_integrity"],"consensus_categories":[],"category_scores_codex":[0.006003833,0.000918674,0.001864088,0.0002056415,0.0003656407,0.0004957885,0.001213719,0.001018173,0.0003057258],"category_scores_gemma":[0.2373653,0.000945504,0.0004676379,0.0007730769,0.0003639985,0.0002606389,0.001455036,0.003209834,0.0001116163],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0005248981,"about_ca_system_score_gemma":0.001053745,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0001609416,"about_ca_topic_score_gemma":0.000001410844,"domain_scores_codex":[0.9915185,0.001788045,0.002460094,0.002220118,0.001084698,0.000928546],"domain_scores_gemma":[0.9705254,0.02227256,0.003376558,0.002135006,0.0007234921,0.0009670057],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"bench_or_experimental","study_design_gemma":"bench_or_experimental","study_design_scores_codex":[0.001458394,0.003546288,0.002498368,0.01215553,0.005127843,0.0005852575,0.0001375539,0.0001388253,0.6211774,0.1683718,0.1846632,0.0001395098],"study_design_scores_gemma":[0.01582103,0.001667111,0.01779351,0.01337547,0.02637359,6.207893e-7,0.0001999013,0.2790303,0.4920601,0.03734736,0.09577196,0.02055908],"study_design_candidate":"bench_or_experimental","study_design_consensus":"bench_or_experimental","genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.07026719,0.001047452,0.8877925,0.0123053,0.01045816,0.005484684,0.007962383,0.004282529,0.0003997493],"genre_scores_gemma":[0.3324022,0.0001748446,0.664225,0.0004289236,0.002192273,0.0002450882,5.724268e-7,0.0003038495,0.00002727883],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.2788914,"threshold_uncertainty_score":0.9992995,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1992216331082018,"score_gpt":0.4269110192685788,"score_spread":0.2276893861603769,"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."}}