{"id":"W2509088678","doi":"10.1101/064626","title":"Whose sample is it anyway? Widespread misannotation of samples in transcriptomics studies","year":2016,"lang":"en","type":"preprint","venue":"bioRxiv (Cold Spring Harbor Laboratory)","topic":"Gene expression and cancer classification","field":"Biochemistry, Genetics and Molecular Biology","cited_by":8,"is_retracted":false,"has_abstract":true,"ca_institutions":"Genome British Columbia; Canada's Michael Smith Genome Sciences Centre; University of British Columbia","funders":"","keywords":"Sample (material); Reliability (semiconductor); Annotation; Set (abstract data type); Computer science; Confidence interval; Data set; Sample size determination; Statistics; Reproducibility; Artificial intelligence; 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":[],"category_scores_codex":[0.1801702,0.00068538,0.001974457,0.005513283,0.002902535,0.006164246,0.002518648,0.002773594,0.001627523],"category_scores_gemma":[0.45066,0.0009789929,0.0009770073,0.007477785,0.007544406,0.004066826,0.004795137,0.002668011,0.0006792743],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001876506,"about_ca_system_score_gemma":0.002143182,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.002377762,"about_ca_topic_score_gemma":0.003912816,"domain_scores_codex":[0.7601888,0.1388567,0.0332825,0.033834,0.03149002,0.002348012],"domain_scores_gemma":[0.4251457,0.3925407,0.07342897,0.07366592,0.03295832,0.002260467],"domain_codex":null,"domain_gemma":"reproducibility","domain_candidate":"reproducibility","domain_consensus":null,"study_design_codex":"observational","study_design_gemma":"observational","study_design_scores_codex":[0.002917367,0.0001771825,0.5925281,0.007175619,0.002523355,0.003953959,0.04725039,0.002267795,0.03310818,0.02593265,0.02506236,0.257103],"study_design_scores_gemma":[0.0002179918,0.0004560939,0.4594321,0.009119235,0.002876914,0.009275703,0.02824035,0.01517993,0.07972568,0.1823452,0.2125621,0.0005686887],"study_design_candidate":"observational","study_design_consensus":"observational","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.6264479,0.02760754,0.2890457,0.03827385,0.004849107,0.0007575172,0.004077717,0.001062867,0.007877869],"genre_scores_gemma":[0.8559673,0.003247794,0.1202087,0.01481681,0.001343405,0.0005621785,0.001707742,0.0006924942,0.001453551],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.9972264,"threshold_uncertainty_score":0.9528423,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.04184005282366721,"score_gpt":0.286059165772216,"score_spread":0.2442191129485488,"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."}}