{"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":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0001650551,0.0002030276,0.000341168,0.00008721307,0.00008651374,0.00003486007,0.0001925275,0.00007199353,0.00003104403],"category_scores_gemma":[0.00005879763,0.0001601722,0.00009741562,0.00005657188,0.000147926,0.000002376586,0.0001019986,0.00009720047,0.000002068047],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00002017098,"about_ca_system_score_gemma":0.00002732918,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00001036369,"about_ca_topic_score_gemma":0.00003184718,"domain_scores_codex":[0.9988605,0.00009743748,0.0004099777,0.0002513243,0.00009839584,0.0002824117],"domain_scores_gemma":[0.9993306,0.00003589356,0.0002304752,0.0001736091,0.00008986097,0.0001395278],"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":[0.0002891344,0.00004750231,0.007083096,0.00001216576,0.00007197377,0.00001850007,0.0001091787,0.00149834,0.9900287,0.000007443191,0.00004498607,0.0007889921],"study_design_scores_gemma":[0.009455054,0.0257601,0.4211029,0.0005168308,0.0005435332,0.001820771,0.0005987731,0.01181262,0.4666802,0.0003538286,0.05847884,0.002876468],"study_design_candidate":"bench_or_experimental","study_design_consensus":"bench_or_experimental","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9957197,0.00162143,0.002007958,0.0003167183,0.0001285126,0.000106578,0.00004122511,0.000001359687,0.00005651854],"genre_scores_gemma":[0.9950901,0.0005733679,0.003733377,0.0003685142,0.0001632023,0.000004593722,0.000004299462,0.00001853473,0.00004398607],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.5233485,"threshold_uncertainty_score":0.6531635,"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."}}