{"id":"W2979310796","doi":"10.1111/2041-210x.13314","title":"MetaCurator: A hidden Markov model‐based toolkit for extracting and curating sequences from taxonomically‐informative genetic markers","year":2019,"lang":"en","type":"article","venue":"Methods in Ecology and Evolution","topic":"Environmental DNA in Biodiversity Studies","field":"Environmental Science","cited_by":54,"is_retracted":false,"has_abstract":true,"ca_institutions":"York University","funders":"Ohio State University","keywords":"Hidden Markov model; Pipeline (software); Sequence database; Biology; Sequence (biology); Computer science; Information retrieval; Artificial intelligence; Genetics; Programming language","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":[0.003962771,0.00321965,0.002471491,0.004882182,0.001216137,0.00243626,0.004080207,0.001507964,0.01595213],"category_scores_gemma":[0.01154223,0.00241912,0.002967525,0.002467252,0.0007328252,0.00239772,0.003741066,0.003562008,0.01667304],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0007516738,"about_ca_system_score_gemma":0.003061165,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.003534956,"about_ca_topic_score_gemma":0.00692724,"domain_scores_codex":[0.9981295,0.0004235527,0.0002661018,0.0006004434,0.0004340066,0.0001463559],"domain_scores_gemma":[0.9965652,0.001788065,0.0006769923,0.0003988283,0.0003876651,0.0001832423],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.003602509,0.0003130221,0.008851791,0.01976262,0.002932243,0.002222845,0.003474062,0.02109701,0.1289097,0.01943414,0.3901132,0.399287],"study_design_scores_gemma":[0.000779192,0.000582572,0.01120255,0.002666022,0.001265882,0.002056533,0.0005210262,0.2086581,0.1060607,0.03657557,0.6287298,0.0009020409],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.008091782,0.001761954,0.5965145,0.0002981107,0.0002795041,0.0006574082,0.09040674,0.2995231,0.002466894],"genre_scores_gemma":[0.02020429,0.0008322436,0.8382649,0.0003308434,0.00007478454,0.002168285,0.1011292,0.03445579,0.002539617],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.01595213,"threshold_uncertainty_score":0.05336523,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02145089154093273,"score_gpt":0.2836678876402798,"score_spread":0.2622169960993471,"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."}}