{"id":"W3162338526","doi":"10.1101/2021.05.17.444395","title":"Hidden viral sequences in public sequencing data and warning for future emerging diseases","year":2021,"lang":"en","type":"preprint","venue":"bioRxiv (Cold Spring Harbor Laboratory)","topic":"Viral Infections and Vectors","field":"Medicine","cited_by":5,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"Institute of Genetics; Ministry of Education, Culture, Sports, Science and Technology; Japan Society for the Promotion of Science; Institute of Medical Science, University of Tokyo; RIKEN; University of Tokyo","keywords":"Biology; Human virome; Virology; Metagenomics; Deep sequencing; Genome; Genetics; Gene","routes":{"ca_aff":false,"ca_fund":true,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.006850799,0.0008200411,0.001060428,0.005256178,0.0008457055,0.002369791,0.0008372233,0.001517254,0.001998506],"category_scores_gemma":[0.01227165,0.000452052,0.0009687773,0.004668383,0.0006734918,0.001786517,0.00193235,0.001732875,0.001295396],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0007108211,"about_ca_system_score_gemma":0.001359826,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00190016,"about_ca_topic_score_gemma":0.003356173,"domain_scores_codex":[0.9954787,0.001258049,0.0006093754,0.001215733,0.001075542,0.0003627192],"domain_scores_gemma":[0.9857607,0.005413244,0.003678356,0.001896702,0.00244011,0.0008108769],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"observational","study_design_gemma":"observational","study_design_scores_codex":[0.001400159,0.0003119863,0.7025623,0.004314743,0.001641113,0.003475682,0.001541012,0.01185948,0.1187167,0.005945873,0.02472437,0.1235067],"study_design_scores_gemma":[0.0001453101,0.0007053664,0.4969241,0.003409829,0.001804285,0.006089103,0.004986773,0.1114284,0.07012498,0.0451198,0.2588756,0.0003863468],"study_design_candidate":"observational","study_design_consensus":"observational","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.7572129,0.02182746,0.08733997,0.004768418,0.001189929,0.0002610394,0.1157829,0.002561045,0.009056442],"genre_scores_gemma":[0.82406,0.003725706,0.07100201,0.001692559,0.0004864404,0.0001649361,0.09750905,0.0002951818,0.001064107],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.006850799,"threshold_uncertainty_score":0.03623092,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.03833287030721555,"score_gpt":0.278002431897346,"score_spread":0.2396695615901305,"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."}}