{"id":"W3133283981","doi":"10.1038/s41467-021-21254-9","title":"Uniform genomic data analysis in the NCI Genomic Data Commons","year":2021,"lang":"en","type":"article","venue":"Nature Communications","topic":"Cancer Genomics and Diagnostics","field":"Biochemistry, Genetics and Molecular Biology","cited_by":148,"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; Epigenomics; Data sharing; Raw data; Computational biology; Genomics; Precision medicine; Computer science; DNA methylation; Genome; Biology; Bioinformatics; Database; Genetics; Gene; 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":[0.04438469,0.001385027,0.002397646,0.009164447,0.002826239,0.009564968,0.007527213,0.00235936,0.02179087],"category_scores_gemma":[0.12468,0.001781417,0.002207844,0.01939118,0.003154735,0.005778348,0.01900952,0.005237915,0.01451242],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.006873402,"about_ca_system_score_gemma":0.02456951,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.04415083,"about_ca_topic_score_gemma":0.02945776,"domain_scores_codex":[0.9685339,0.00648862,0.00524919,0.005339575,0.01238965,0.001999078],"domain_scores_gemma":[0.8986471,0.01629587,0.004143104,0.05906809,0.01745383,0.00439207],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"not_applicable","study_design_gemma":"not_applicable","study_design_scores_codex":[0.001292394,0.0002536114,0.01735906,0.001395576,0.0003765319,0.001182056,0.002160636,0.007162503,0.005840568,0.08007924,0.7901366,0.09276121],"study_design_scores_gemma":[0.0004153934,0.00006693629,0.00989218,0.0006844837,0.00008638565,0.0003724918,0.0006087015,0.009316775,0.006960576,0.04902424,0.922412,0.0001596803],"study_design_candidate":"not_applicable","study_design_consensus":"not_applicable","genre_codex":"dataset","genre_gemma":"methods","genre_scores_codex":[0.008053591,0.0008815105,0.1848658,0.008523047,0.00111541,0.002808274,0.6828779,0.07604176,0.0348327],"genre_scores_gemma":[0.03415534,0.0006590438,0.1657374,0.003074608,0.0002819228,0.0041266,0.7771366,0.009918104,0.004910384],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.04438469,"threshold_uncertainty_score":0.2347315,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.04565861463920134,"score_gpt":0.3347831478506129,"score_spread":0.2891245332114116,"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."}}