{"id":"W2558624268","doi":"10.1371/journal.pone.0167047","title":"Simulating Next-Generation Sequencing Datasets from Empirical Mutation and Sequencing Models","year":2016,"lang":"en","type":"article","venue":"PLoS ONE","topic":"Cancer Genomics and Diagnostics","field":"Biochemistry, Genetics and Molecular Biology","cited_by":130,"is_retracted":false,"has_abstract":true,"ca_institutions":"Ontario Institute for Cancer Research","funders":"Ontario Institute for Cancer Research; National Science Foundation","keywords":"Computer science; Benchmarking; Software; Set (abstract data type); Data mining; Genome; DNA sequencing; Scripting language; Computational biology; Biology; Genetics; Gene","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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.003565898,0.0006576813,0.0005919365,0.0008081702,0.000494587,0.0009596655,0.002064091,0.001472793,0.00466379],"category_scores_gemma":[0.01411302,0.0006779446,0.001204462,0.001187243,0.0006774799,0.001289189,0.0009513979,0.001916092,0.001107387],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001200304,"about_ca_system_score_gemma":0.001327264,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.01022246,"about_ca_topic_score_gemma":0.01393611,"domain_scores_codex":[0.9991216,0.0003705127,0.00006044048,0.0001889274,0.0001788045,0.00007974751],"domain_scores_gemma":[0.9938151,0.004717328,0.0001975611,0.000700518,0.0003689137,0.0002005636],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"simulation_or_modeling","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.0001609854,0.00005995698,0.007873737,0.0001503634,0.00008442606,0.0001344401,0.0001675529,0.9657259,0.002051819,0.01191638,0.004007977,0.007666571],"study_design_scores_gemma":[0.0000328038,0.00002163666,0.0008335572,0.00001280637,0.00001423764,0.00003734757,0.00002748925,0.9857603,0.001560957,0.009594971,0.002085441,0.00001844151],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.4343781,0.0006118701,0.5169846,0.001144992,0.000314731,0.0004993864,0.02059134,0.01618719,0.009287798],"genre_scores_gemma":[0.7090929,0.0004676566,0.2581407,0.0006444721,0.00005419233,0.001247664,0.02304009,0.002060609,0.00525175],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.01022246,"threshold_uncertainty_score":0.0203259,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.160761244021002,"score_gpt":0.2840296338623527,"score_spread":0.1232683898413507,"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."}}