{"id":"W3159161400","doi":"10.1101/2021.04.29.442004","title":"VirusTaxo: Taxonomic classification of virus genome using multi-class hierarchical classification by k-mer enrichment","year":2021,"lang":"en","type":"preprint","venue":"bioRxiv (Cold Spring Harbor Laboratory)","topic":"Bacteriophages and microbial interactions","field":"Environmental Science","cited_by":1,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of British Columbia","funders":"","keywords":"Virus classification; Biological classification; Genome; Taxonomic rank; Contig; Biology; Computational biology; DNA sequencing; Genus; Genetics; DNA; Evolutionary biology; Gene; Taxon; Botany","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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.001612947,0.00146888,0.00107163,0.002584603,0.001129031,0.001460322,0.001204634,0.001216284,0.00316331],"category_scores_gemma":[0.00239084,0.0005493899,0.00186371,0.001123112,0.000490425,0.001409356,0.001354318,0.001698085,0.002158078],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001200995,"about_ca_system_score_gemma":0.001243398,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.004613766,"about_ca_topic_score_gemma":0.005250051,"domain_scores_codex":[0.9990606,0.0001419465,0.00006530451,0.0003900009,0.000214477,0.0001276824],"domain_scores_gemma":[0.9990281,0.0003022473,0.0001147361,0.000159585,0.0002907899,0.0001046014],"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":[0.001943235,0.0008877231,0.03872334,0.001054791,0.0006998543,0.0003362114,0.0005167555,0.03695276,0.3176911,0.00533155,0.04365705,0.5522056],"study_design_scores_gemma":[0.00007206153,0.0002592334,0.008936282,0.00005626596,0.0001017278,0.0001601032,0.0001210245,0.8766156,0.09278461,0.007234855,0.0135643,0.00009385412],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.1816467,0.001133141,0.7334566,0.000688988,0.0004061967,0.0008495769,0.01231673,0.06593152,0.003570568],"genre_scores_gemma":[0.2502447,0.0002817161,0.7158244,0.0003978745,0.00009001853,0.0007114189,0.0247209,0.002063794,0.00566525],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.004613766,"threshold_uncertainty_score":0.01058233,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.03105426594171763,"score_gpt":0.246490706445138,"score_spread":0.2154364405034204,"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."}}