{"id":"W2110019612","doi":"10.1093/bioinformatics/btp478","title":"Methods for labeling error detection in microarrays based on the effect of data perturbation on the regression model","year":2009,"lang":"en","type":"article","venue":"Bioinformatics","topic":"Gene expression and cancer classification","field":"Biochemistry, Genetics and Molecular Biology","cited_by":27,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"University of British Columbia","keywords":"Support vector machine; Computer science; Artificial intelligence; Pattern recognition (psychology); Algorithm; Regression; Data mining; DNA microarray; Source code; Mathematics; Statistics","routes":{"ca_aff":false,"ca_fund":true,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.001032918,0.00009859433,0.00008549158,0.00004639134,0.00008838944,0.00001553553,0.0002740276,0.00009537557,0.000001779444],"category_scores_gemma":[0.000429314,0.00004897948,0.0000429061,0.0000927763,0.00002019517,0.00000808523,0.00002615757,0.00007519184,8.869999e-7],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00001606761,"about_ca_system_score_gemma":0.00003087179,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":5.888787e-7,"about_ca_topic_score_gemma":0.000001510247,"domain_scores_codex":[0.9993392,0.0001225586,0.0002225943,0.0001234899,0.00009607782,0.00009608559],"domain_scores_gemma":[0.9990089,0.0001478419,0.0001632625,0.0006266696,0.00003690571,0.00001643764],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"bench_or_experimental","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.0004069076,0.00002721392,0.000009851535,0.00003666958,0.000005170396,1.065734e-8,0.0001050425,0.005984309,0.8575261,0.0001250397,0.001854409,0.1339193],"study_design_scores_gemma":[0.0002353922,0.0002895392,0.00003206404,0.00003811812,0.00000571435,1.301553e-7,0.00003710699,0.5381966,0.4600022,0.00006681235,0.001057968,0.00003837036],"study_design_candidate":"bench_or_experimental","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.1312902,0.0001284037,0.863404,0.002541195,0.000149183,0.001187029,0.00004040277,0.00001445753,0.001245067],"genre_scores_gemma":[0.9692321,0.00002368196,0.02954443,0.0008906884,0.00003750162,0.00004714945,0.0001476766,0.000007763649,0.00006902937],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.8379419,"threshold_uncertainty_score":0.1997325,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.05680105849511773,"score_gpt":0.3634192508253875,"score_spread":0.3066181923302698,"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."}}