{"id":"W4403152041","doi":"10.1101/2024.10.04.615514","title":"Defining a tandem repeat catalog and variation clusters for genome-wide analyses","year":2024,"lang":"en","type":"preprint","venue":"bioRxiv (Cold Spring Harbor Laboratory)","topic":"Forensic and Genetic Research","field":"Biochemistry, Genetics and Molecular Biology","cited_by":13,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of British Columbia; Canada's Michael Smith Genome Sciences Centre","funders":"","keywords":"Variation (astronomy); Genome; Population; Computer science; Database; Computational biology; Information retrieval; Biology; Genetics; Medicine","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":[],"consensus_categories":[],"category_scores_codex":[0.005795942,0.0009029327,0.001362944,0.01163543,0.001595383,0.003802574,0.002452607,0.001069476,0.00741686],"category_scores_gemma":[0.01247538,0.0006418732,0.001040807,0.01971864,0.0006568731,0.003000666,0.00298811,0.001788624,0.006528215],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001349866,"about_ca_system_score_gemma":0.004471176,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.008423422,"about_ca_topic_score_gemma":0.009247392,"domain_scores_codex":[0.9968521,0.0004317382,0.0005007787,0.0009852217,0.0009677825,0.0002624202],"domain_scores_gemma":[0.9899357,0.001711574,0.001239293,0.002456193,0.003873531,0.0007837251],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.0009325604,0.0005017251,0.1097492,0.002753258,0.0002996986,0.001615718,0.002622389,0.01278134,0.05643066,0.07878941,0.1609945,0.5725296],"study_design_scores_gemma":[0.0003860702,0.0005572794,0.1770153,0.001320299,0.0003797044,0.002358594,0.002025087,0.05190278,0.02380297,0.06130446,0.6784251,0.0005222893],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.1055143,0.002584367,0.6898555,0.001199894,0.0005805285,0.004092937,0.1431816,0.02164487,0.03134602],"genre_scores_gemma":[0.05407201,0.0008792285,0.7798353,0.0003523868,0.0002082575,0.00196548,0.1562855,0.002108248,0.004293462],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.01163543,"threshold_uncertainty_score":0.03065223,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01920968624667846,"score_gpt":0.2772954510363662,"score_spread":0.2580857647896877,"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."}}