{"id":"W2978660451","doi":"10.1101/788919","title":"Uniform Genomic Data Analysis in the NCI Genomic Data Commons","year":2019,"lang":"en","type":"preprint","venue":"bioRxiv (Cold Spring Harbor Laboratory)","topic":"Cancer Genomics and Diagnostics","field":"Biochemistry, Genetics and Molecular Biology","cited_by":20,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"National Cancer Institute; National Institutes of Health; University of Texas MD Anderson Cancer Center; Canada's Michael Smith Genome Sciences Centre; U.S. Department of Health and Human Services","keywords":"Workflow; Data sharing; Raw data; Epigenomics; Genomics; Computational biology; Precision medicine; Copy-number variation; DNA methylation; Biology; Computer science; Genome; Database; Gene; Genetics; Medicine; Gene expression","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":[],"consensus_categories":[],"category_scores_codex":[0.03953415,0.001395498,0.002127489,0.008155921,0.002971417,0.01018031,0.006080853,0.002243262,0.01287173],"category_scores_gemma":[0.1195302,0.001681986,0.002265726,0.01734954,0.003326702,0.005210455,0.01739399,0.004770536,0.009886878],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.005629606,"about_ca_system_score_gemma":0.01881742,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.03585938,"about_ca_topic_score_gemma":0.02203017,"domain_scores_codex":[0.9609044,0.008636077,0.005103412,0.007155699,0.0162618,0.001938604],"domain_scores_gemma":[0.8756243,0.02023384,0.004211323,0.07656462,0.01935361,0.004012367],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"not_applicable","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.001504002,0.0003168161,0.02421577,0.001216984,0.0006189261,0.001796519,0.001900708,0.01578928,0.0082188,0.1252825,0.7127362,0.1064034],"study_design_scores_gemma":[0.0005039415,0.00007524789,0.01488616,0.0006841249,0.0001405505,0.0005785366,0.000828871,0.03485666,0.01451947,0.1074199,0.8252953,0.0002112492],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"dataset","genre_gemma":"methods","genre_scores_codex":[0.01517074,0.000872669,0.2926452,0.009197501,0.001150649,0.001727512,0.5369923,0.1152295,0.02701388],"genre_scores_gemma":[0.07507807,0.0005400414,0.2034947,0.002486454,0.0003189506,0.002308874,0.6995021,0.011994,0.004276928],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.03953415,"threshold_uncertainty_score":0.2090791,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02850292568356836,"score_gpt":0.2536023526968867,"score_spread":0.2250994270133183,"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."}}