{"id":"W3127141800","doi":"10.1101/2021.02.01.429279","title":"The impact of distributional assumptions in gene-set and pathway analysis: how far can it go wrong?","year":2021,"lang":"en","type":"preprint","venue":"bioRxiv (Cold Spring Harbor Laboratory)","topic":"Gene expression and cancer classification","field":"Biochemistry, Genetics and Molecular Biology","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"School of Medicine, New York University; York University","keywords":"Normality; Parametric statistics; Set (abstract data type); Multivariate statistics; Multivariate normal distribution; Multivariate analysis; Computer science; Expression (computer science); Univariate; Multiple comparisons problem; Computational biology; Econometrics; Data mining; Statistics; Biology; Mathematics; Machine learning","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":"metacan-v3-hybrid-931329e0061c","candidate_categories":["metaresearch"],"consensus_categories":[],"category_scores_codex":[0.1391565,0.0009958622,0.002560105,0.002025069,0.001415024,0.004488293,0.004135663,0.002904169,0.001345932],"category_scores_gemma":[0.4125695,0.0006116542,0.002820323,0.002882421,0.007334601,0.009320455,0.004210285,0.006646828,0.0005251737],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00203535,"about_ca_system_score_gemma":0.002536489,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.002729154,"about_ca_topic_score_gemma":0.00199651,"domain_scores_codex":[0.9119372,0.06382386,0.003829191,0.006053188,0.01342891,0.0009276961],"domain_scores_gemma":[0.48895,0.4410889,0.01539892,0.03484906,0.01764537,0.002067715],"domain_codex":null,"domain_gemma":"methods","domain_candidate":"methods","domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.003136418,0.0005441902,0.1883575,0.002159166,0.003486395,0.00145462,0.002550291,0.2629269,0.006935606,0.1332601,0.009924475,0.3852643],"study_design_scores_gemma":[0.0001287522,0.0007307293,0.02944145,0.0009485201,0.0003143574,0.0008560523,0.001471675,0.5749172,0.008032926,0.377219,0.005639835,0.0002994802],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.1334754,0.006815848,0.8402936,0.01444695,0.0008688968,0.0001867714,0.0004336329,0.000780866,0.00269798],"genre_scores_gemma":[0.8471688,0.00193737,0.1456775,0.002721751,0.000616321,0.0003329514,0.0005987573,0.0003819091,0.000564536],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.8608435,"threshold_uncertainty_score":0.7359388,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01552098160064649,"score_gpt":0.2542113856946583,"score_spread":0.2386904040940118,"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."}}