{"id":"W4213220107","doi":"10.1016/j.xpro.2022.101175","title":"TreeTuner: A pipeline for minimizing redundancy and complexity in large phylogenetic datasets","year":2022,"lang":"en","type":"article","venue":"STAR Protocols","topic":"Genomics and Phylogenetic Studies","field":"Biochemistry, Genetics and Molecular Biology","cited_by":4,"is_retracted":false,"has_abstract":true,"ca_institutions":"McGill University; Western University; Dalhousie University","funders":"Natural Sciences and Engineering Research Council of Canada; Gordon and Betty Moore Foundation","keywords":"Phylogenetic tree; Pipeline (software); Redundancy (engineering); Trimming; Computer science; Multiple sequence alignment; Data mining; Protocol (science); Computational biology; Biology; Sequence alignment; 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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.00714587,0.003103069,0.002666575,0.004125663,0.00340743,0.003886012,0.004140003,0.001656617,0.01510991],"category_scores_gemma":[0.01439215,0.002838482,0.00333215,0.004053709,0.001301588,0.004503852,0.005422472,0.006452261,0.01860516],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001061878,"about_ca_system_score_gemma":0.003921402,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.002428257,"about_ca_topic_score_gemma":0.005828953,"domain_scores_codex":[0.9969715,0.0005712398,0.000320179,0.001147399,0.0007348118,0.0002548668],"domain_scores_gemma":[0.9959163,0.001663249,0.0004898608,0.0009892591,0.0006464972,0.000294847],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"not_applicable","study_design_scores_codex":[0.001668681,0.00043309,0.009529036,0.005135217,0.001731642,0.0007718949,0.003510417,0.00665866,0.2489859,0.01230654,0.2984609,0.410808],"study_design_scores_gemma":[0.0009701794,0.0007787644,0.0323013,0.0009020731,0.00109012,0.002570938,0.001215699,0.2052499,0.1876739,0.07762383,0.4885047,0.001118523],"study_design_candidate":"not_applicable","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"software","genre_scores_codex":[0.01385604,0.0007301975,0.7988604,0.0003799819,0.0003235035,0.0007443416,0.03223468,0.1510183,0.001852616],"genre_scores_gemma":[0.01552955,0.0003278269,0.916927,0.0003040505,0.00006370159,0.001385718,0.04155109,0.02245295,0.0014581],"genre_candidate":"software","genre_consensus":null,"teacher_disagreement_score":0.01510991,"threshold_uncertainty_score":0.05054766,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.05669721992346096,"score_gpt":0.3277885316308824,"score_spread":0.2710913117074215,"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."}}