{"id":"W4411720712","doi":"10.1101/2025.06.23.661092","title":"Comparative performance of reference-based metagenomic tools to identify species-level taxa among families of bacteria: benchmarking <i>Mycobacteriaceae</i> and <i>Neisseriaceae</i>","year":2025,"lang":"en","type":"preprint","venue":"bioRxiv (Cold Spring Harbor Laboratory)","topic":"Genomics and Phylogenetic Studies","field":"Biochemistry, Genetics and Molecular Biology","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"Institut National de la Recherche Scientifique; McGill University Health Centre","funders":"Canadian Institutes of Health Research; Alliance de recherche numérique du Canada; Natural Sciences and Engineering Research Council of Canada; Ministère de la Santé","keywords":"Benchmarking; Metagenomics; Taxon; Neisseriaceae; Biology; Bacteria; Geography; Ecology; Paleontology; Genetics; Business","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":[0.003889395,0.0009790433,0.0007487854,0.002910711,0.0004339461,0.001082612,0.001164995,0.000952695,0.0005292519],"category_scores_gemma":[0.007450209,0.0003774042,0.0009158734,0.001737825,0.0004965019,0.0008595044,0.001017436,0.0004723945,0.0003935343],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0004326614,"about_ca_system_score_gemma":0.0003249838,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.002145653,"about_ca_topic_score_gemma":0.002072214,"domain_scores_codex":[0.9964508,0.0009612662,0.0003176604,0.0009515831,0.001119615,0.0001991697],"domain_scores_gemma":[0.9952731,0.002285241,0.0004992131,0.0006189531,0.00109481,0.0002286661],"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":[0.004690091,0.001133514,0.1903571,0.001614369,0.001493965,0.0003340343,0.0008869846,0.09268954,0.5842557,0.0009384631,0.001592095,0.1200143],"study_design_scores_gemma":[0.0001027995,0.003841891,0.1219833,0.00007797994,0.0003437822,0.0005245484,0.0004996996,0.3683783,0.5008894,0.0007912585,0.002419972,0.000147097],"study_design_candidate":"bench_or_experimental","study_design_consensus":"bench_or_experimental","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9822296,0.001098496,0.0133079,0.00005341312,0.00003014798,0.00004800664,0.001566506,0.0008572159,0.0008087057],"genre_scores_gemma":[0.9521363,0.000342659,0.041314,0.00004392431,0.0000148156,0.00009567728,0.005665447,0.0001316854,0.0002555071],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.003889395,"threshold_uncertainty_score":0.02056932,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.03247631467002048,"score_gpt":0.2489922285502265,"score_spread":0.216515913880206,"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."}}