{"id":"W2884303974","doi":"10.1111/1755-0998.12926","title":"Supervised machine learning outperforms taxonomy‐based environmental <scp>DNA</scp> metabarcoding applied to biomonitoring","year":2018,"lang":"en","type":"article","venue":"Molecular Ecology Resources","topic":"Environmental DNA in Biodiversity Studies","field":"Environmental Science","cited_by":165,"is_retracted":false,"has_abstract":true,"ca_institutions":"Cytodiagnostics (Canada)","funders":"Schweizerischer Nationalfonds zur Förderung der Wissenschaftlichen Forschung; Deutsche Forschungsgemeinschaft; Carl-Zeiss-Stiftung","keywords":"Biology; Environmental DNA; Taxonomic rank; Bioindicator; Biological classification; Amplicon; Biodiversity; Taxonomy (biology); Taxon; Computational biology; Ecology; Machine learning; Evolutionary biology; Computer science; Genetics; Polymerase chain reaction","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":[0.004398304,0.001053204,0.0007241695,0.0008847256,0.0003781714,0.0008751375,0.0007901607,0.0007712208,0.001085883],"category_scores_gemma":[0.008733586,0.0002026615,0.0008423305,0.000474996,0.000415936,0.001110955,0.0005935257,0.0009654426,0.0006713411],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.000855202,"about_ca_system_score_gemma":0.001159101,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.005483896,"about_ca_topic_score_gemma":0.005895463,"domain_scores_codex":[0.9979135,0.001044304,0.0001365433,0.0004764973,0.0003009322,0.0001280909],"domain_scores_gemma":[0.99389,0.003960236,0.0006523627,0.0006024466,0.0007122727,0.0001825815],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"simulation_or_modeling","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.0006461409,0.0007443976,0.07103352,0.0003337031,0.0006248867,0.0001142876,0.0001202343,0.5764704,0.005590713,0.001182844,0.003378918,0.3397599],"study_design_scores_gemma":[0.000007785206,0.0001864667,0.005206003,0.00002011586,0.00002599044,0.00002031908,0.00001482964,0.9915193,0.001913665,0.0007496337,0.0003237662,0.00001216343],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.856252,0.002574244,0.1294609,0.001137596,0.0002296787,0.000115695,0.0009719631,0.002848472,0.006409588],"genre_scores_gemma":[0.9781857,0.0002120576,0.01919591,0.0001450882,0.0000534363,0.00003857758,0.001110625,0.0000565691,0.001002142],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.005483896,"threshold_uncertainty_score":0.02326071,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01276858083435116,"score_gpt":0.1961043850239808,"score_spread":0.1833358041896297,"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."}}