{"id":"W1565378970","doi":"10.1038/nmeth.3407","title":"Combining tumor genome simulation with crowdsourcing to benchmark somatic single-nucleotide-variant detection","year":2015,"lang":"en","type":"article","venue":"Nature Methods","topic":"Cancer Genomics and Diagnostics","field":"Biochemistry, Genetics and Molecular Biology","cited_by":369,"is_retracted":false,"has_abstract":true,"ca_institutions":"Ontario Institute for Cancer Research","funders":"National Cancer Institute; Natural Sciences and Engineering Research Council of Canada; Prostate Cancer Canada; National Human Genome Research Institute; Ontario Institute for Cancer Research; Mater Foundation; Agence Nationale de la Recherche; Government of Ontario; Google; Canadian Institutes of Health Research; Genome Canada; National Institutes of Health; Movember Foundation","keywords":"Benchmark (surveying); False positive paradox; Genome; Somatic cell; Benchmarking; Computational biology; In silico; Computer science; Biology; Human genome; Mutation; Pipeline (software); Genetics; Gene; Machine learning","routes":{"ca_aff":true,"ca_fund":true,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":false},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0006915013,0.0001709125,0.0001808275,0.00007944466,0.00008671488,0.00006855858,0.0001325106,0.0001922292,0.000004935295],"category_scores_gemma":[0.0008788024,0.00015653,0.00004966828,0.0002019415,0.00002367902,0.000005662722,0.00009625491,0.0002203897,0.000004169574],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00008281984,"about_ca_system_score_gemma":0.00008725296,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00002046722,"about_ca_topic_score_gemma":0.00005985008,"domain_scores_codex":[0.9989189,0.0001467775,0.0001992513,0.0003502862,0.0001503269,0.0002344568],"domain_scores_gemma":[0.9990665,0.0001120623,0.0001087978,0.0003318271,0.0001939167,0.0001868449],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"bench_or_experimental","study_design_gemma":"bench_or_experimental","study_design_scores_codex":[0.0002631203,0.00005712896,0.0003577029,0.00002298879,0.00005402684,0.00001446036,0.0003494481,0.07423285,0.9081396,0.00004298928,0.00008451565,0.01638118],"study_design_scores_gemma":[0.002650379,0.003001537,0.005956677,0.0001041094,0.0001905144,0.0001468102,0.0004541783,0.02300728,0.8038028,0.001400139,0.1581598,0.001125758],"study_design_candidate":"bench_or_experimental","study_design_consensus":"bench_or_experimental","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.5777326,0.0007343062,0.4195933,0.00008330004,0.0005544173,0.0002488639,0.000007412317,0.00002137862,0.001024362],"genre_scores_gemma":[0.8374904,0.000003291385,0.1612478,0.0006959679,0.0004367392,0.00001399946,0.00003165274,0.0000332753,0.0000468528],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.2597578,"threshold_uncertainty_score":0.6383108,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01846596755281029,"score_gpt":0.3194236580117129,"score_spread":0.3009576904589026,"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."}}