{"id":"W2190667587","doi":"10.6339/jds.2003.01(2).131","title":"Analysis of Unbalanced Microarray Data","year":2021,"lang":"en","type":"article","venue":"Journal of Data Science","topic":"Gene expression and cancer classification","field":"Biochemistry, Genetics and Molecular Biology","cited_by":5,"is_retracted":false,"has_abstract":true,"ca_institutions":"McGill University","funders":"Natural Sciences and Engineering Research Council of Canada; National Institutes of Health","keywords":"Microarray analysis techniques; China; Data science; Microarray databases; Library science; Data analysis; Data mining; Computer science; Statistics; Political science; Mathematics; Biology; Gene","routes":{"ca_aff":true,"ca_fund":true,"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":[0.01198696,0.000930177,0.001916281,0.003560282,0.0007696779,0.001474328,0.001053352,0.000520326,0.009725481],"category_scores_gemma":[0.0558451,0.0005125732,0.001417651,0.005044945,0.0007450304,0.001541007,0.001111669,0.00170937,0.002122036],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0008531461,"about_ca_system_score_gemma":0.001364187,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0009644263,"about_ca_topic_score_gemma":0.001539128,"domain_scores_codex":[0.9892367,0.003787236,0.0008543865,0.002137707,0.003532005,0.0004519845],"domain_scores_gemma":[0.9616882,0.02443079,0.002138702,0.007262041,0.00405883,0.0004215067],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.002038458,0.0002330799,0.05451868,0.001778278,0.001759607,0.0007590309,0.0003797228,0.03888746,0.05127212,0.03018928,0.1117336,0.7064508],"study_design_scores_gemma":[0.0003149345,0.0007670636,0.1261126,0.0002929922,0.0008684893,0.001773123,0.0005567023,0.4924067,0.02773083,0.2018335,0.1471051,0.0002379782],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.04990222,0.002030298,0.9076702,0.001709415,0.0009671418,0.0004377561,0.03177967,0.002787235,0.00271608],"genre_scores_gemma":[0.3161926,0.002130138,0.5758262,0.0008427283,0.001232248,0.002854455,0.09364902,0.001469995,0.005802597],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.01198696,"threshold_uncertainty_score":0.06339383,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.06929950204773849,"score_gpt":0.3692050382421703,"score_spread":0.2999055361944318,"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."}}