{"id":"W6901900520","doi":"10.6084/m9.figshare.12417881.v3","title":"EukProt: a database of genome-scale predicted proteins across the diversity of eukaryotes","year":2022,"lang":"en","type":"dataset","venue":"Figshare","topic":"","field":"","cited_by":2,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Alberta","funders":"","keywords":"Phylogenomics; Identifier; Phylogenetic tree; Sequence database; Set (abstract data type); Web server; Genomics; Tree (set theory); Biological database; Genome","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":[0.001144776,0.002890291,0.001953722,0.006560799,0.001738383,0.003091644,0.00203122,0.001250187,0.07118239],"category_scores_gemma":[0.00290706,0.001757124,0.00119031,0.008241303,0.0003004591,0.003685476,0.003222204,0.001750728,0.1089914],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0006046687,"about_ca_system_score_gemma":0.00127487,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.001001115,"about_ca_topic_score_gemma":0.001621179,"domain_scores_codex":[0.9990682,0.0001148993,0.0001443903,0.0002787648,0.000262535,0.0001312744],"domain_scores_gemma":[0.9990128,0.0001605971,0.0002353011,0.0001304385,0.0002759244,0.0001848676],"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":[0.001007898,0.00004632242,0.001510112,0.004000617,0.0001134649,0.0003837637,0.0001830332,0.000305847,0.0334321,0.00167795,0.9235231,0.03381577],"study_design_scores_gemma":[0.0002032564,0.0001483213,0.008517867,0.0007278707,0.0001280259,0.001273683,0.0002321365,0.001000036,0.01056295,0.002512653,0.974533,0.0001603144],"study_design_candidate":"not_applicable","study_design_consensus":"not_applicable","genre_codex":"dataset","genre_gemma":"dataset","genre_scores_codex":[0.007828152,0.003725546,0.0124656,0.0005745147,0.0005033512,0.0001265582,0.9153963,0.04741222,0.01196766],"genre_scores_gemma":[0.00491415,0.0009925667,0.01463439,0.0001711306,0.0000738642,0.0001268977,0.9698201,0.00600324,0.003263754],"genre_candidate":"dataset","genre_consensus":"dataset","teacher_disagreement_score":0.07118239,"threshold_uncertainty_score":0.2381288,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.05363692601495947,"score_gpt":0.2856649561066191,"score_spread":0.2320280300916596,"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."}}