{"id":"W4309326199","doi":"10.1101/2022.11.17.513905","title":"Maximizing the reliability and the number of species assignments in metabarcoding studies","year":2022,"lang":"en","type":"preprint","venue":"bioRxiv (Cold Spring Harbor Laboratory)","topic":"Environmental DNA in Biodiversity Studies","field":"Environmental Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"Fisheries and Oceans Canada","funders":"","keywords":"Barcode; Reliability (semiconductor); Ranking (information retrieval); Taxon; Biodiversity; Global biodiversity; Biology; Taxonomic rank; DNA barcoding; Environmental DNA; Ecology; Computer science; Information retrieval","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":["metaresearch"],"consensus_categories":[],"category_scores_codex":[0.04002857,0.001348584,0.001484446,0.004690812,0.001310272,0.002627206,0.001522148,0.001394122,0.0007789888],"category_scores_gemma":[0.119264,0.001067026,0.00125711,0.003462678,0.001225121,0.003096575,0.002988456,0.001279618,0.0006326194],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0007100871,"about_ca_system_score_gemma":0.001187928,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00127858,"about_ca_topic_score_gemma":0.002816045,"domain_scores_codex":[0.9709873,0.01633652,0.003318939,0.004879083,0.003769627,0.000708496],"domain_scores_gemma":[0.8827889,0.07301014,0.01663656,0.01004314,0.01632719,0.001194068],"domain_codex":null,"domain_gemma":"methods","domain_candidate":"methods","domain_consensus":null,"study_design_codex":"observational","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.002181231,0.0003867745,0.4303241,0.004277058,0.001873182,0.0006719003,0.00434802,0.04452908,0.2481077,0.005398471,0.002325231,0.2555772],"study_design_scores_gemma":[0.0002073887,0.001730993,0.3534005,0.001359105,0.002733031,0.001719848,0.002445057,0.2696816,0.323015,0.02145344,0.02170695,0.0005470424],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"methods","genre_scores_codex":[0.6233665,0.003039988,0.3675847,0.0003879645,0.0001106624,0.0004291041,0.001076987,0.001524752,0.00247937],"genre_scores_gemma":[0.64618,0.0005189404,0.3508737,0.000113583,0.00005218531,0.0002824273,0.001246915,0.0003228309,0.0004094297],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.9599714,"threshold_uncertainty_score":0.2116938,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02526826030976441,"score_gpt":0.2339516658927092,"score_spread":0.2086834055829448,"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."}}