{"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":"codex-gemma-dda1882f352a","candidate_categories":["sts"],"consensus_categories":[],"category_scores_codex":[0.001335073,0.00009162056,0.0001465069,0.00002222334,0.001861924,0.00001015023,0.0008324081,0.00002788134,0.00003341648],"category_scores_gemma":[0.0000936389,0.00007315881,0.0001101534,0.00009464103,0.0002142015,0.000001287375,0.003819452,0.0005009029,1.706562e-7],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00001476626,"about_ca_system_score_gemma":0.00006731563,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0005391635,"about_ca_topic_score_gemma":0.0001923506,"domain_scores_codex":[0.9987687,0.0004779522,0.0002251902,0.00007591112,0.0002838898,0.0001683533],"domain_scores_gemma":[0.998922,0.00003248405,0.0002242587,0.0005047566,0.0002705583,0.00004595147],"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":[0.0003077234,0.0005133757,0.1158049,0.00004770671,0.000357336,0.000002550099,0.01235773,0.0009144074,0.8677469,0.00001200938,0.001255962,0.0006794266],"study_design_scores_gemma":[0.002979332,0.002827049,0.7549299,0.00003086131,0.0001826208,0.0003764998,0.03592134,0.0001121072,0.1642279,0.0008670237,0.03707115,0.0004741525],"study_design_candidate":"bench_or_experimental","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9969668,0.0009180513,0.0003050207,0.0003082376,0.00007938438,0.0001962872,0.0009840183,0.000001308731,0.0002408883],"genre_scores_gemma":[0.9990045,0.0001494513,0.0004696735,0.00007055894,0.00005910967,0.00001179272,0.00007310192,0.00000821541,0.000153605],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.703519,"threshold_uncertainty_score":0.9994375,"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."}}