{"id":"W4390590654","doi":"10.1099/mgen.0.001165","title":"An imputed ancestral reference genome for the Mycobacterium tuberculosis complex better captures structural genomic diversity for reference-based alignment workflows","year":2024,"lang":"en","type":"article","venue":"Microbial Genomics","topic":"Genomics and Phylogenetic Studies","field":"Biochemistry, Genetics and Molecular Biology","cited_by":11,"is_retracted":false,"has_abstract":true,"ca_institutions":"McGill University; Institut National de la Recherche Scientifique","funders":"Fonds de Recherche du Québec - Santé; Huck Institutes of the Life Sciences; Canadian Institutes of Health Research; Alliance de recherche numérique du Canada; U.S. Department of Agriculture; Canada Research Chairs; Department for International Development; Bill and Melinda Gates Foundation","keywords":"Mycobacterium tuberculosis complex; Reference genome; Biology; Genome; Genetics; Sequence (biology); Comparative genomics; Whole genome sequencing; Computational biology; Evolutionary biology; Genomics; Mycobacterium tuberculosis; Tuberculosis; Gene","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":["metaepi_narrow"],"consensus_categories":[],"category_scores_codex":[0.000187896,0.0003547164,0.0002688094,0.00004625544,0.0006647319,0.0001923953,0.0006005944,0.0002002172,0.00002117989],"category_scores_gemma":[0.000007106388,0.0002880764,0.0002276206,0.00005944938,0.0001605066,0.000003711963,0.0003257741,0.0001190667,0.000004270677],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001097152,"about_ca_system_score_gemma":0.0001715602,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.000145382,"about_ca_topic_score_gemma":0.0003260194,"domain_scores_codex":[0.9983314,0.00004298488,0.000315075,0.0007063748,0.00007851945,0.0005256763],"domain_scores_gemma":[0.9991437,0.00005345832,0.00008486277,0.0004908447,0.0001266248,0.0001004905],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"bench_or_experimental","study_design_gemma":"not_applicable","study_design_scores_codex":[0.000294116,0.00002029855,0.0004503849,0.00006835082,0.0002974948,9.384556e-7,0.0002365904,0.00196282,0.9927532,0.00008233496,0.001414513,0.002418927],"study_design_scores_gemma":[0.003071618,0.00172952,0.09343236,0.00003682142,0.0008550927,0.00003328061,0.0003622878,0.009702098,0.1563385,0.001072802,0.731382,0.001983722],"study_design_candidate":"bench_or_experimental","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9880561,0.002038301,0.004397593,0.0005246772,0.0006428485,0.001057725,0.003240111,0.00001803191,0.00002466471],"genre_scores_gemma":[0.987735,0.0002544113,0.008177584,0.001001666,0.000984914,0.00009858482,0.001634161,0.0000533398,0.00006033677],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.8364148,"threshold_uncertainty_score":0.9999571,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.03431125821683473,"score_gpt":0.261999938002015,"score_spread":0.2276886797851803,"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."}}