{"id":"W4297888171","doi":"10.24072/pcjournal.173","title":"EukProt: A database of genome-scale predicted proteins across the diversity of eukaryotes","year":2022,"lang":"en","type":"article","venue":"Peer Community Journal","topic":"Genomics and Phylogenetic Studies","field":"Biochemistry, Genetics and Molecular Biology","cited_by":212,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Alberta","funders":"Science for Life Laboratory; Fundación Bancaria Caixa d'Estalvis i Pensions de Barcelona; Svenska Forskningsrådet Formas; Deutsche Forschungsgemeinschaft; European Commission; Gordon and Betty Moore Foundation; Agence Nationale de la Recherche; Vetenskapsrådet; Generalitat de Catalunya; National Science Foundation","keywords":"Diversity (politics); Genome; Biology; Scale (ratio); Evolutionary biology; Computational biology; Geography; Genetics; Gene; Sociology; Cartography; Anthropology","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":[0.001448424,0.002447305,0.00186386,0.007829916,0.002124537,0.003338441,0.001875704,0.001045217,0.01704783],"category_scores_gemma":[0.003192333,0.001423227,0.001146428,0.01001043,0.0003536658,0.003353917,0.002853826,0.001471775,0.01885603],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0004921737,"about_ca_system_score_gemma":0.001482934,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0008150704,"about_ca_topic_score_gemma":0.001325239,"domain_scores_codex":[0.9989625,0.000130956,0.0001845965,0.000366004,0.0002389827,0.0001169829],"domain_scores_gemma":[0.9986175,0.0002445377,0.0003489628,0.000207017,0.0002971298,0.0002848488],"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":[0.007646552,0.0004253886,0.01363275,0.01272934,0.000848571,0.002340873,0.001378875,0.002441902,0.2396336,0.005352417,0.5679153,0.1456544],"study_design_scores_gemma":[0.0007545743,0.0007569903,0.04910022,0.001373032,0.000695335,0.005380412,0.0008398507,0.004880344,0.04259603,0.006904683,0.8862742,0.0004443359],"study_design_candidate":"not_applicable","study_design_consensus":"not_applicable","genre_codex":"dataset","genre_gemma":"dataset","genre_scores_codex":[0.06242688,0.006737051,0.02696855,0.0006005584,0.0004184528,0.0002100675,0.8629982,0.02996907,0.009671109],"genre_scores_gemma":[0.01553428,0.001368603,0.02894188,0.0001010468,0.00007002059,0.0001396145,0.9503741,0.001894136,0.001576288],"genre_candidate":"dataset","genre_consensus":"dataset","teacher_disagreement_score":0.01704783,"threshold_uncertainty_score":0.05703062,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.0274859170767314,"score_gpt":0.2588532816666173,"score_spread":0.2313673645898859,"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."}}