{"id":"W4385975493","doi":"10.1371/journal.pone.0283536","title":"MT-MAG: Accurate and interpretable machine learning for complete or partial taxonomic assignments of metagenome-assembled genomes","year":2023,"lang":"en","type":"article","venue":"PLoS ONE","topic":"Genomics and Phylogenetic Studies","field":"Biochemistry, Genetics and Molecular Biology","cited_by":7,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Waterloo","funders":"Vector Institute; Natural Sciences and Engineering Research Council of Canada; Government of Canada; Canadian Institute for Advanced Research","keywords":"Metagenomics; Weighting; Taxonomic rank; Genome; Artificial intelligence; Pattern recognition (psychology); Biology; Biological classification; Feature (linguistics); Phylogenetic tree; Computer science; Computational biology; Machine learning; Genetics; Evolutionary biology; Physics; Ecology; Gene; Taxon","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":[0.002381357,0.002017887,0.001071163,0.002281378,0.0006076759,0.00202471,0.002485518,0.001431649,0.003821201],"category_scores_gemma":[0.01003601,0.0007198965,0.001867793,0.001275578,0.0005991043,0.002564336,0.002093511,0.001935757,0.002689323],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0007225457,"about_ca_system_score_gemma":0.001005318,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.001497919,"about_ca_topic_score_gemma":0.001931451,"domain_scores_codex":[0.9985916,0.0002866552,0.0001135388,0.0004716441,0.0004385187,0.00009801374],"domain_scores_gemma":[0.997556,0.001171038,0.0002969088,0.0004692028,0.0003800492,0.0001267721],"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":[0.001740482,0.0003108332,0.02085414,0.001438357,0.0007549008,0.0004727936,0.0007369172,0.1406955,0.06968977,0.01661496,0.0822192,0.6644722],"study_design_scores_gemma":[0.00007189293,0.0001010041,0.001892051,0.00006504965,0.00004322959,0.0001152036,0.00005963739,0.9329903,0.03215865,0.01582251,0.01660918,0.00007130571],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.03538207,0.0005716227,0.7766071,0.0004033651,0.0002459152,0.000103965,0.006662629,0.1780627,0.001960578],"genre_scores_gemma":[0.1835043,0.0002624479,0.7954457,0.0004300868,0.0001082549,0.0003706106,0.01327725,0.005061178,0.001540068],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.003821201,"threshold_uncertainty_score":0.01278317,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.08676248436545862,"score_gpt":0.2636243748884113,"score_spread":0.1768618905229527,"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."}}