{"id":"W4391897827","doi":"10.1093/bioinformatics/btae066","title":"VariantDetective: an accurate all-in-one pipeline for detecting consensus bacterial SNPs and SVs","year":2024,"lang":"en","type":"article","venue":"Bioinformatics","topic":"Genomics and Phylogenetic Studies","field":"Biochemistry, Genetics and Molecular Biology","cited_by":10,"is_retracted":false,"has_abstract":true,"ca_institutions":"Canadian Food Inspection Agency","funders":"Canadian Food Inspection Agency","keywords":"MIT License; Single-nucleotide polymorphism; Source code; Computational biology; Benchmarking; Pipeline (software); SNP; Biology; Computer science; Genetics; Data mining; Software; Gene; Genotype; Programming language","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":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0002897926,0.0001422236,0.0001463763,0.00005177439,0.00006715087,0.00008381488,0.00007692277,0.000108556,0.000003013866],"category_scores_gemma":[0.0001130897,0.0001335481,0.00004138707,0.00005319,0.00004885556,0.000002561051,0.00008528848,0.00005893645,0.000002507003],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00001153849,"about_ca_system_score_gemma":0.0000572021,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00001129939,"about_ca_topic_score_gemma":0.00007855654,"domain_scores_codex":[0.9992093,0.00001759121,0.0002946194,0.0001800787,0.00005071777,0.0002476945],"domain_scores_gemma":[0.9996262,0.00004740631,0.00005310693,0.0001602296,0.0000493689,0.00006373562],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"bench_or_experimental","study_design_gemma":"bench_or_experimental","study_design_scores_codex":[0.0002745234,0.0000502337,0.0002337967,0.0004167218,0.0002066211,0.000004288776,0.001576402,0.0001228617,0.9200897,0.000312563,0.0005176561,0.07619462],"study_design_scores_gemma":[0.00611037,0.00311925,0.007226913,0.0002493477,0.0003603496,0.0002628677,0.003651674,0.350972,0.4342487,0.002498139,0.189207,0.002093419],"study_design_candidate":"bench_or_experimental","study_design_consensus":"bench_or_experimental","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9906201,0.0009006367,0.007056083,0.0001129342,0.0004219276,0.0003815554,0.0001437347,0.00001210377,0.0003509649],"genre_scores_gemma":[0.9846733,0.0002457978,0.01452061,0.00009865082,0.0003064644,0.00002832309,0.0000524061,0.00001925108,0.00005526577],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.4858409,"threshold_uncertainty_score":0.5445932,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.03652047725925867,"score_gpt":0.285410330898393,"score_spread":0.2488898536391343,"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."}}