{"id":"W4243426183","doi":"10.12688/f1000research.9471.2","title":"Whose sample is it anyway? Widespread misannotation of samples in transcriptomics studies","year":2016,"lang":"en","type":"preprint","venue":"F1000Research","topic":"Gene expression and cancer classification","field":"Biochemistry, Genetics and Molecular Biology","cited_by":3,"is_retracted":false,"has_abstract":true,"ca_institutions":"Genome British Columbia; Canada's Michael Smith Genome Sciences Centre; University of British Columbia","funders":"National Institute of General Medical Sciences; National Institutes of Health","keywords":"Sample (material); Annotation; Reliability (semiconductor); Confidence interval; Sample size determination; Statistics; Set (abstract data type); Reproducibility; Data set; Computer science; Biology; Bioinformatics; Mathematics","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":["metaresearch","research_integrity"],"consensus_categories":["metaresearch"],"category_scores_codex":[0.2568305,0.00088783,0.002809886,0.006456705,0.003974472,0.00733072,0.003307153,0.003966618,0.001756757],"category_scores_gemma":[0.5587714,0.00148681,0.001375337,0.009431508,0.01092389,0.0065138,0.006183826,0.003741853,0.0009644102],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.002716157,"about_ca_system_score_gemma":0.003570479,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.004049346,"about_ca_topic_score_gemma":0.006014353,"domain_scores_codex":[0.6964318,0.1716668,0.04762915,0.04116908,0.04052312,0.002580108],"domain_scores_gemma":[0.3853849,0.4232846,0.0712921,0.07967322,0.03791491,0.002450363],"domain_codex":null,"domain_gemma":"methods","domain_candidate":"methods","domain_consensus":null,"study_design_codex":"observational","study_design_gemma":"observational","study_design_scores_codex":[0.003578312,0.0001795762,0.4284899,0.01214699,0.002857348,0.004412765,0.08029573,0.001730238,0.02715274,0.03612614,0.04096058,0.3620697],"study_design_scores_gemma":[0.0003383673,0.0005193945,0.3423471,0.01335794,0.002894583,0.009280274,0.03601644,0.00856269,0.0461387,0.2242014,0.3156642,0.0006790122],"study_design_candidate":"observational","study_design_consensus":"observational","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.49339,0.05376327,0.3305009,0.0921653,0.010858,0.001373485,0.005380803,0.001376842,0.01119137],"genre_scores_gemma":[0.7179831,0.008400707,0.2201812,0.04200001,0.003379087,0.001526923,0.002733276,0.001147023,0.002648718],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.9960334,"threshold_uncertainty_score":0.9164603,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1718735376716192,"score_gpt":0.4305569874046562,"score_spread":0.258683449733037,"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."}}