{"id":"W4405221092","doi":"10.1093/bioinformatics/btae719","title":"PheWAS analysis on large-scale biobank data with PheTK","year":2024,"lang":"en","type":"article","venue":"Bioinformatics","topic":"Genetic Associations and Epidemiology","field":"Biochemistry, Genetics and Molecular Biology","cited_by":13,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Toronto","funders":"National Human Genome Research Institute; National Institutes of Health","keywords":"Biobank; Computer science; Scale (ratio); Data science; Bioinformatics; Biology; Geography; Cartography","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":[0.008498127,0.001645594,0.001411981,0.003621724,0.001593879,0.004867549,0.00224649,0.000987289,0.04532903],"category_scores_gemma":[0.04161498,0.001132053,0.002784028,0.004736164,0.001101612,0.003335664,0.005319322,0.002415081,0.02890729],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001003391,"about_ca_system_score_gemma":0.005361785,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00580242,"about_ca_topic_score_gemma":0.008301069,"domain_scores_codex":[0.9944699,0.001310103,0.000657972,0.001873748,0.001220357,0.0004680038],"domain_scores_gemma":[0.9832622,0.008578945,0.001214465,0.003997562,0.002383824,0.0005630355],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"not_applicable","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.003983064,0.0003578328,0.05105815,0.004702367,0.002838789,0.001414976,0.001497497,0.02058643,0.009768601,0.01230458,0.7448202,0.1466674],"study_design_scores_gemma":[0.001687909,0.0005870672,0.07548153,0.001104771,0.001209169,0.001920015,0.001156049,0.1459782,0.02403506,0.08993114,0.6562217,0.0006872825],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"dataset","genre_gemma":"methods","genre_scores_codex":[0.04189236,0.001124215,0.2422474,0.002792381,0.001075274,0.001319056,0.4217338,0.2700148,0.01780071],"genre_scores_gemma":[0.1313537,0.000753988,0.2241227,0.00230401,0.0003036477,0.005126603,0.5680502,0.05542142,0.0125636],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.04532903,"threshold_uncertainty_score":0.1516407,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01748673113263469,"score_gpt":0.2820495558834093,"score_spread":0.2645628247507746,"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."}}