{"id":"W2044610621","doi":"10.1128/jb.184.8.2072-2080.2002","title":"Inferring Genome Trees by Using a Filter To Eliminate Phylogenetically Discordant Sequences and a Distance Matrix Based on Mean Normalized BLASTP Scores","year":2002,"lang":"en","type":"article","venue":"Journal of Bacteriology","topic":"Genomics and Phylogenetic Studies","field":"Biochemistry, Genetics and Molecular Biology","cited_by":122,"is_retracted":false,"has_abstract":true,"ca_institutions":"Canadian Institute for Advanced Research; University of Ottawa","funders":"","keywords":"Biology; Genome; ORFS; Tree of life (biology); Genetics; Tree (set theory); Evolutionary biology; Bacterial genome size; Similarity (geometry); Computational biology; Gene; Phylogenetics; Open reading frame; Peptide sequence; Artificial intelligence","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":[0.00521996,0.001233685,0.001392334,0.006183424,0.001685918,0.001981319,0.001397487,0.001329885,0.001716097],"category_scores_gemma":[0.01805099,0.0006542691,0.001700511,0.003882296,0.0006323948,0.001370318,0.001054502,0.001778904,0.001125645],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0007951008,"about_ca_system_score_gemma":0.002227734,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00359773,"about_ca_topic_score_gemma":0.009038848,"domain_scores_codex":[0.9980531,0.0003975142,0.0002460987,0.0006444783,0.0005236525,0.0001352972],"domain_scores_gemma":[0.9933991,0.003704655,0.0007732863,0.0004668786,0.001341721,0.0003143013],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.001491099,0.000873216,0.1014359,0.001006043,0.001100523,0.0009524139,0.002207299,0.02253268,0.2060023,0.007571117,0.003663942,0.6511636],"study_design_scores_gemma":[0.0003464128,0.001414035,0.09955595,0.0002258562,0.0008662521,0.002822889,0.001299477,0.7577766,0.09378904,0.02225662,0.01930871,0.0003382036],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.299048,0.0003156067,0.6934699,0.0001662203,0.00005484228,0.0003601642,0.001256343,0.004348346,0.0009807164],"genre_scores_gemma":[0.2077968,0.0001226764,0.7880429,0.00005942862,0.00002211119,0.0002063896,0.003029499,0.000281103,0.0004389443],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.006183424,"threshold_uncertainty_score":0.02760613,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01532073216114388,"score_gpt":0.2486588886696093,"score_spread":0.2333381565084654,"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."}}