{"id":"W4402073580","doi":"10.20944/preprints202408.1963.v1","title":"How Trustworthy Are Genomic Sequences of SARS-CoV-2 in GenBank?","year":2024,"lang":"en","type":"preprint","venue":"Preprints.org","topic":"SARS-CoV-2 and COVID-19 Research","field":"Medicine","cited_by":2,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Ottawa","funders":"Natural Sciences and Engineering Research Council of Canada","keywords":"GenBank; Genome; Annotation; Biology; Phylogenetic tree; Genetics; Genome project; Computational biology; Genomics; 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":["metaresearch"],"consensus_categories":[],"category_scores_codex":[0.03290353,0.0006065866,0.001490218,0.01215163,0.001257452,0.007771306,0.001608275,0.002527227,0.004565271],"category_scores_gemma":[0.21006,0.0007839436,0.000575696,0.01079457,0.002862265,0.009562776,0.00197442,0.001521364,0.00613512],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001595667,"about_ca_system_score_gemma":0.003507753,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.004792485,"about_ca_topic_score_gemma":0.002761076,"domain_scores_codex":[0.9646195,0.01570149,0.007283607,0.004331669,0.007120381,0.0009433723],"domain_scores_gemma":[0.8407814,0.08280098,0.02069278,0.02652878,0.0267492,0.002446909],"domain_codex":null,"domain_gemma":"reproducibility","domain_candidate":"reproducibility","domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"observational","study_design_scores_codex":[0.003272489,0.0002971849,0.3471107,0.004677294,0.0007267501,0.004607163,0.01501474,0.006526021,0.03125789,0.03631146,0.06329101,0.4869072],"study_design_scores_gemma":[0.0003218415,0.0007992942,0.2338812,0.009554929,0.001405924,0.008713501,0.03251669,0.03165178,0.03367295,0.0769678,0.5697639,0.0007501588],"study_design_candidate":"observational","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.5562042,0.04158411,0.1770368,0.1019113,0.00662169,0.0006365923,0.06613716,0.005531806,0.04433627],"genre_scores_gemma":[0.7659138,0.01513768,0.1571751,0.00644675,0.001686928,0.0003035034,0.04836521,0.001472343,0.003498565],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.9670964,"threshold_uncertainty_score":0.1740126,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.2099241529328065,"score_gpt":0.4152517786440248,"score_spread":0.2053276257112183,"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."}}