{"id":"W3009070426","doi":"10.1099/mgen.0.000341","title":"An assessment of genome annotation coverage across the bacterial tree of life","year":2020,"lang":"en","type":"article","venue":"Microbial Genomics","topic":"Genomics and Phylogenetic Studies","field":"Biochemistry, Genetics and Molecular Biology","cited_by":122,"is_retracted":false,"has_abstract":true,"ca_institutions":"Wilfrid Laurier University; University of Waterloo","funders":"Natural Sciences and Engineering Research Council of Canada","keywords":"Annotation; Genome; Proteome; Biology; Bacterial genome size; Computational biology; Homology (biology); Genome project; Gene Annotation; Gene; Phylogenetics; Genome size; Genetics","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":[],"consensus_categories":[],"category_scores_codex":[0.0001340724,0.0001249287,0.0002034809,0.000008526953,0.00007714543,0.0000186258,0.0002800596,0.00007715866,0.00001076735],"category_scores_gemma":[0.00002232826,0.0001075187,0.00009456344,0.00005173776,0.0001141013,0.000001103758,0.0001523284,0.00005088378,0.000001150274],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00001009062,"about_ca_system_score_gemma":0.0001652696,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00002624531,"about_ca_topic_score_gemma":0.00004854261,"domain_scores_codex":[0.9991859,0.00005388817,0.0003162609,0.0002265917,0.00005763028,0.0001597914],"domain_scores_gemma":[0.9993808,0.000008775916,0.0001907162,0.0002577527,0.00009959571,0.00006237053],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"bench_or_experimental","study_design_gemma":"bench_or_experimental","study_design_scores_codex":[0.000144844,0.00004468815,0.001399617,0.00002028733,0.0000866534,2.072891e-7,0.0008357694,0.002071832,0.9943848,0.00002870645,0.0001029109,0.000879632],"study_design_scores_gemma":[0.002210707,0.00151088,0.1530937,0.000004316058,0.00007056874,0.000003602561,0.0006821336,0.0004430819,0.7839783,0.00004170352,0.05755097,0.0004099775],"study_design_candidate":"bench_or_experimental","study_design_consensus":"bench_or_experimental","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9970818,0.0002546796,0.001480268,0.0001949578,0.0001411289,0.0002024601,0.0005144147,0.000001709088,0.0001286218],"genre_scores_gemma":[0.9976469,0.000246258,0.001076261,0.0004086227,0.0004158456,0.000004366937,0.0001711768,0.00001889782,0.00001164417],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.2104065,"threshold_uncertainty_score":0.4384484,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01321447599047545,"score_gpt":0.2681903846297509,"score_spread":0.2549759086392755,"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."}}