{"id":"W3103395583","doi":"10.1101/2020.05.22.111344","title":"Optimizing predictive models to prioritize viral discovery in zoonotic reservoirs","year":2020,"lang":"en","type":"preprint","venue":"bioRxiv (Cold Spring Harbor Laboratory)","topic":"Viral Infections and Vectors","field":"Medicine","cited_by":23,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Toronto; Université de Montréal","funders":"Institut de Valorisation des Données; Vlaamse regering; Fonds Wetenschappelijk Onderzoek; Wellcome Trust; National Science Foundation","keywords":"Host (biology); Computer science; Biology; Ecology","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":"codex-gemma-dda1882f352a","candidate_categories":["metaepi_narrow"],"consensus_categories":[],"category_scores_codex":[0.0004326782,0.0006817565,0.001084491,0.0006577915,0.0001306916,0.0002694399,0.0003848557,0.0005932462,0.0000231086],"category_scores_gemma":[0.000329695,0.000720024,0.0002957654,0.001008425,0.0001002859,0.0005125083,0.0008582088,0.001695207,0.00006046233],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0009771643,"about_ca_system_score_gemma":0.0009520419,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0003790711,"about_ca_topic_score_gemma":0.00001587975,"domain_scores_codex":[0.9963913,0.0001410301,0.0007942771,0.00141236,0.0005102121,0.0007508075],"domain_scores_gemma":[0.997552,0.00006477438,0.0002364994,0.001206171,0.0003296383,0.0006109303],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"bench_or_experimental","study_design_gemma":"observational","study_design_scores_codex":[0.002321917,0.001654277,0.039256,0.003522064,0.001272072,0.001772763,0.0006382392,0.1501133,0.7937047,0.00381807,0.001921216,0.000005356684],"study_design_scores_gemma":[0.01062545,0.004048793,0.4816726,0.01581367,0.002098028,4.643672e-7,0.000202781,0.2352679,0.2393717,0.0001815714,0.00387085,0.006846243],"study_design_candidate":"bench_or_experimental","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9658278,0.0005137655,0.02801309,0.001385352,0.00118065,0.002173341,0.0002608886,0.0005458692,0.00009931155],"genre_scores_gemma":[0.9888754,0.0001500439,0.008945279,0.0006513417,0.0007669189,0.0004019009,9.982539e-7,0.0001911916,0.00001688096],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.554333,"threshold_uncertainty_score":0.9995251,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.03179665096273403,"score_gpt":0.2541684972931359,"score_spread":0.2223718463304019,"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."}}