{"id":"W4394675060","doi":"10.1093/bioinformatics/btae192","title":"VirusPredictor: XGBoost-based software to predict virus-related sequences in human data","year":2024,"lang":"en","type":"article","venue":"Bioinformatics","topic":"Genomics and Phylogenetic Studies","field":"Biochemistry, Genetics and Molecular Biology","cited_by":13,"is_retracted":false,"has_abstract":true,"ca_institutions":"Centre Hospitalier Universitaire Sainte-Justine","funders":"National Institute of Allergy and Infectious Diseases; National Institutes of Health; Open Medicine Foundation; U.S. Department of Defense","keywords":"Python (programming language); Contig; Genome; Metagenomics; Biology; Sanger sequencing; Computational biology; Software; Genetics; DNA sequencing; Genomics; Human genome; Computer science; Gene","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":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0002868773,0.0001903857,0.0001614583,0.0001197738,0.00008217991,0.00007902732,0.0005366797,0.0001411446,0.00001421692],"category_scores_gemma":[0.00009205863,0.0001704337,0.00005156581,0.0001986639,0.00007654005,0.000004976534,0.0004034101,0.0001036444,0.00004396577],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00002438586,"about_ca_system_score_gemma":0.0001533796,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00005119099,"about_ca_topic_score_gemma":0.0002789621,"domain_scores_codex":[0.9988081,0.0000207224,0.0004396292,0.000291622,0.0001612073,0.0002787606],"domain_scores_gemma":[0.9991243,0.00002246232,0.00004822171,0.0006724118,0.00003516318,0.00009741526],"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":[0.0001580135,0.0002859791,0.03227,0.001221122,0.0008301729,0.00006906211,0.003464165,0.007920874,0.74764,0.000739225,0.1491951,0.05620632],"study_design_scores_gemma":[0.001796907,0.00173343,0.01702236,0.0005836024,0.0001723662,0.00003433744,0.0007881384,0.0546087,0.09199524,0.0006216945,0.8289309,0.001712286],"study_design_candidate":"not_applicable","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.986547,0.003298883,0.004346248,0.0002558729,0.0009115345,0.0005846029,0.001734052,0.00006447258,0.002257332],"genre_scores_gemma":[0.9898593,0.000174473,0.008233069,0.000467045,0.0001370246,0.00002746279,0.0008694417,0.00002814626,0.0002040021],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.6797358,"threshold_uncertainty_score":0.6950083,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02375274263784608,"score_gpt":0.274237705772119,"score_spread":0.2504849631342729,"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."}}