{"id":"W4394432483","doi":"10.6084/m9.figshare.19681987","title":"Additional file 3 of DeepRepeat: direct quantification of short tandem repeats on signal data from nanopore sequencing","year":2022,"lang":"en","type":"dataset","venue":"Open MIND","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; Computational biology; Microsatellite; Computer science; Biology; Genetics; DNA sequencing; DNA; Genome; Base sequence; Gene","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":[0.002189295,0.002128944,0.001786702,0.002558456,0.0009499539,0.002614416,0.002868067,0.002239162,0.4392553],"category_scores_gemma":[0.01219149,0.0008055568,0.001305887,0.004198614,0.0005796327,0.001688079,0.001584021,0.001815093,0.13218],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001411717,"about_ca_system_score_gemma":0.002036492,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.009425055,"about_ca_topic_score_gemma":0.01898622,"domain_scores_codex":[0.9988371,0.0001902435,0.0001443894,0.0004366245,0.0002156835,0.0001759723],"domain_scores_gemma":[0.9928908,0.004613979,0.0005439502,0.0007421409,0.0008763596,0.0003327149],"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":[0.0002241441,0.0000663343,0.002762717,0.002880404,0.0001027179,0.00005042099,0.00004077115,0.0008012009,0.0002791266,0.0006713137,0.9890271,0.003093725],"study_design_scores_gemma":[0.002107293,0.0001043306,0.01297554,0.001260608,0.0002179026,0.0002334316,0.0001499472,0.00129737,0.001340752,0.006960714,0.9732512,0.0001010684],"study_design_candidate":"not_applicable","study_design_consensus":"not_applicable","genre_codex":"dataset","genre_gemma":"dataset","genre_scores_codex":[0.00005127681,0.00001179231,0.00006757561,0.0000161558,0.000004407176,0.000008131736,0.999572,0.0001370864,0.0001316015],"genre_scores_gemma":[0.0008889455,0.0000339691,0.0005620689,0.00008556119,0.000007889572,0.0001728489,0.9973513,0.0002248749,0.0006726037],"genre_candidate":"dataset","genre_consensus":"dataset","teacher_disagreement_score":0.4392553,"threshold_uncertainty_score":0.7998343,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.07699100003676569,"score_gpt":0.3044801268558257,"score_spread":0.22748912681906,"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."}}