{"id":"W4386545312","doi":"10.1101/2023.09.07.556366","title":"An imputed ancestral reference genome for the <i>Mycobacterium tuberculosis</i> complex better captures structural genomic diversity for reference-based alignment workflows","year":2023,"lang":"en","type":"preprint","venue":"bioRxiv (Cold Spring Harbor Laboratory)","topic":"Genomics and Phylogenetic Studies","field":"Biochemistry, Genetics and Molecular Biology","cited_by":1,"is_retracted":false,"has_abstract":true,"ca_institutions":"McGill University; Institut National de la Recherche Scientifique","funders":"Huck Institutes of the Life Sciences; Alliance de recherche numérique du Canada; Department for International Development; Bill and Melinda Gates Foundation; U.S. Department of Agriculture","keywords":"Mycobacterium tuberculosis complex; Reference genome; Biology; Genome; Sequence (biology); Whole genome sequencing; Sequence analysis; Genetics; Computational biology; 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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.002240905,0.0006130414,0.0007871831,0.0008700561,0.00107496,0.001338312,0.001257992,0.000969322,0.004285375],"category_scores_gemma":[0.006281042,0.000517541,0.0009905046,0.001486967,0.0004527092,0.0008745522,0.001113386,0.00167651,0.002253196],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0009878851,"about_ca_system_score_gemma":0.001276612,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.003444084,"about_ca_topic_score_gemma":0.004973163,"domain_scores_codex":[0.9985207,0.0003068815,0.0001325628,0.0005419535,0.0003719923,0.0001259571],"domain_scores_gemma":[0.9972687,0.0008156734,0.0003545351,0.0005551738,0.0008377929,0.0001682436],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"bench_or_experimental","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.00290133,0.0005293913,0.02910794,0.001358491,0.000239473,0.0008153814,0.001102964,0.0337756,0.8114501,0.01291434,0.01390458,0.09190044],"study_design_scores_gemma":[0.0002841479,0.001385087,0.06364343,0.0003333219,0.0004639024,0.001271565,0.0004866781,0.170111,0.6107704,0.009641085,0.1412565,0.0003528574],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"methods","genre_scores_codex":[0.5926032,0.0008935398,0.33338,0.0008134978,0.0004914687,0.0004082528,0.05290341,0.01116382,0.007342796],"genre_scores_gemma":[0.4194981,0.0002681775,0.45432,0.0004203409,0.00007222299,0.0003404649,0.1196364,0.003266834,0.002177532],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.004285375,"threshold_uncertainty_score":0.01433599,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.04222164552642018,"score_gpt":0.250452611260807,"score_spread":0.2082309657343868,"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."}}