{"id":"W6913037803","doi":"10.5524/100977","title":"Supporting data for \"Future-proofing and maximizing the utility of metadata: The PHA4GE SARS-CoV-2 contextual data specification package\"","year":2022,"lang":"en","type":"dataset","venue":"GigaDB","topic":"","field":"","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"Biotechnology and Biological Sciences Research Council; Fundação para a Ciência e a Tecnologia; Genome Canada","keywords":"Interoperability; Consistency (knowledge bases); Contextual design; Standardization; Data integration; Data aggregator; Data quality; Data consistency; Alliance","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":"codex-gemma-dda1882f352a","candidate_categories":["metaepi_narrow","open_science"],"consensus_categories":["open_science"],"category_scores_codex":[0.01452773,0.000750861,0.001118613,0.0001831405,0.001208768,0.0006567592,0.01192209,0.0002477106,0.0008320768],"category_scores_gemma":[0.003903697,0.0004978807,0.0001505213,0.0006210323,0.000737752,0.002061904,0.0111095,0.001443423,0.00003797023],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00008951875,"about_ca_system_score_gemma":0.0004146601,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0008590318,"about_ca_topic_score_gemma":0.001838856,"domain_scores_codex":[0.9927531,0.001029151,0.001748363,0.002332789,0.001237731,0.0008988155],"domain_scores_gemma":[0.9764266,0.001970293,0.002943001,0.01838632,0.0001994609,0.00007430273],"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":[0.0002310237,0.000112199,0.00001663354,0.0003782817,0.0004897079,0.000008446376,0.0001573706,2.130982e-7,0.001618259,0.0001265573,0.991926,0.004935297],"study_design_scores_gemma":[0.0007444226,0.00004214783,0.0001991699,0.00005313845,0.001343402,0.00004882685,0.002845003,0.0009339714,0.0009384977,0.0001380688,0.9922179,0.0004954275],"study_design_candidate":"not_applicable","study_design_consensus":"not_applicable","genre_codex":"dataset","genre_gemma":"dataset","genre_scores_codex":[0.0002466888,0.004063546,0.0006638939,0.0008400097,0.0009325907,0.002992408,0.9901428,0.00007748609,0.00004063787],"genre_scores_gemma":[0.001605072,0.0003658462,0.0006255066,0.0003289342,0.001191808,0.0001880937,0.9954941,0.0001647966,0.00003586664],"genre_candidate":"dataset","genre_consensus":"dataset","teacher_disagreement_score":0.01632653,"threshold_uncertainty_score":0.9997473,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.2391469643514842,"score_gpt":0.4007605994116636,"score_spread":0.1616136350601794,"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."}}