{"id":"W4394520474","doi":"10.6084/m9.figshare.19681984","title":"Additional file 2 of DeepRepeat: direct quantification of short tandem repeats on signal data from nanopore sequencing","year":2022,"lang":"en","type":"dataset","venue":"Figshare","topic":"Genomics and Phylogenetic Studies","field":"Biochemistry, Genetics and Molecular Biology","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Manitoba","funders":"","keywords":"Nanopore sequencing; Tandem repeat; Direct repeat; Nanopore; Computational biology; SIGNAL (programming language); Computer science; Biology; DNA sequencing; Genetics; Nanotechnology; DNA; Genome; Materials science; Base sequence; Programming language","routes":{"ca_aff":true,"ca_fund":false,"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":["insufficient_payload"],"consensus_categories":[],"category_scores_codex":[0.001880384,0.002126629,0.001832597,0.002611891,0.001009507,0.002667818,0.002940905,0.002141006,0.5259149],"category_scores_gemma":[0.01103822,0.0008189221,0.001275532,0.004414426,0.0005499154,0.001927289,0.001597741,0.001756902,0.1648135],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001299206,"about_ca_system_score_gemma":0.00201044,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.007913237,"about_ca_topic_score_gemma":0.0163937,"domain_scores_codex":[0.9990399,0.0001439651,0.0001105843,0.0003817859,0.0001751501,0.0001484847],"domain_scores_gemma":[0.9935503,0.00416227,0.0004285161,0.0007092191,0.0007932899,0.0003562972],"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":[0.000189029,0.00005982634,0.001971503,0.00240593,0.00007069846,0.00004274915,0.00003256352,0.0006035629,0.0002120288,0.000618482,0.9907933,0.003000321],"study_design_scores_gemma":[0.001984515,0.00010375,0.01106328,0.001129054,0.0001687767,0.0002131019,0.0001383666,0.001181052,0.001127618,0.007093657,0.9757013,0.00009535308],"study_design_candidate":"not_applicable","study_design_consensus":"not_applicable","genre_codex":"dataset","genre_gemma":"dataset","genre_scores_codex":[0.00004369276,0.00001036798,0.00005980556,0.000016203,0.00000451416,0.000008549998,0.9995701,0.0001408456,0.000145959],"genre_scores_gemma":[0.0007812977,0.00003240771,0.0005707007,0.00007806672,0.000008474533,0.0001730482,0.9973876,0.00025783,0.0007105572],"genre_candidate":"dataset","genre_consensus":"dataset","teacher_disagreement_score":0.5259149,"threshold_uncertainty_score":0.6762248,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.07373905943946615,"score_gpt":0.2751022244139436,"score_spread":0.2013631649744774,"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."}}