{"id":"W3019597618","doi":"10.1186/s13059-020-02017-z","title":"ExpansionHunter Denovo: a computational method for locating known and novel repeat expansions in short-read sequencing data","year":2020,"lang":"en","type":"article","venue":"Genome biology","topic":"Genomics and Phylogenetic Studies","field":"Biochemistry, Genetics and Molecular Biology","cited_by":227,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Toronto; BC Children's Hospital; SickKids Foundation; University of British Columbia; Hospital for Sick Children; Canada Research Chairs","funders":"National Health and Medical Research Council; Canadian Institutes of Health Research","keywords":"Computational biology; Biology; Genome; Computer science; R package; 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":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0002549126,0.0001557571,0.0002251815,0.00003660388,0.00009698805,0.00001216249,0.0002483802,0.0001233896,0.000004929987],"category_scores_gemma":[0.0001712058,0.0001473474,0.00003559964,0.00006925742,0.00007466105,0.000001519215,0.0005593151,0.00006782699,0.000001241272],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00001397965,"about_ca_system_score_gemma":0.00009301144,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00005443851,"about_ca_topic_score_gemma":0.00009288447,"domain_scores_codex":[0.998725,0.00004968227,0.0002820008,0.000655763,0.00003504462,0.000252532],"domain_scores_gemma":[0.9994397,0.00008026703,0.00004111249,0.0002926484,0.00006713663,0.0000791456],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"bench_or_experimental","study_design_gemma":"not_applicable","study_design_scores_codex":[0.00004991101,0.0000156925,0.002727731,0.00002069299,0.00005808821,9.311286e-7,0.0004731963,0.001265701,0.9914349,0.0001858313,0.00009307276,0.003674241],"study_design_scores_gemma":[0.01500966,0.007123473,0.1222151,0.0001768972,0.0005260484,0.0005534912,0.01526902,0.2834288,0.07947497,0.006303083,0.4643895,0.005529878],"study_design_candidate":"bench_or_experimental","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.7124861,0.003070386,0.2827617,0.0007461712,0.00008343946,0.0003354109,0.0004368902,0.000005588025,0.00007432258],"genre_scores_gemma":[0.8927995,0.0001191806,0.1049303,0.001066817,0.0002404608,0.00003897194,0.0007721685,0.00001909578,0.00001352736],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.9119599,"threshold_uncertainty_score":0.6008652,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.09981444176931657,"score_gpt":0.3401248156679272,"score_spread":0.2403103738986106,"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."}}