{"id":"W3163273456","doi":"10.1186/s12859-021-04180-x","title":"Profile hidden Markov model sequence analysis can help remove putative pseudogenes from DNA barcoding and metabarcoding datasets","year":2021,"lang":"en","type":"article","venue":"BMC Bioinformatics","topic":"Environmental DNA in Biodiversity Studies","field":"Environmental Science","cited_by":45,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Guelph","funders":"Government of Canada; Ontario Genomics; Genome Canada","keywords":"Pseudogene; Barcode; Amplicon; Computational biology; DNA barcoding; Biology; Gene; Sequence analysis; Genetics; DNA sequencing; Coding region; Genome; Computer science; Polymerase chain reaction; Evolutionary biology","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":["metaepi_narrow"],"consensus_categories":[],"category_scores_codex":[0.0002087402,0.0002594694,0.0003461701,0.00007137043,0.0003876349,0.00009994179,0.0002770296,0.00007763349,0.0004122403],"category_scores_gemma":[0.00009584797,0.0002501366,0.0001086806,0.0004841405,0.0003645149,0.0006272368,0.001318667,0.0001281118,0.0001942235],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0002523358,"about_ca_system_score_gemma":0.0000151518,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0002942093,"about_ca_topic_score_gemma":0.0003291855,"domain_scores_codex":[0.9983391,0.00006094152,0.0003715082,0.0003886219,0.0004837081,0.0003561773],"domain_scores_gemma":[0.9990897,0.0001334995,0.0001897681,0.0004164948,0.000009668765,0.0001608346],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"observational","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.00003042723,0.0001081474,0.9401416,0.0001347341,0.001016661,0.00006142856,0.009314369,0.009245438,0.01913853,0.00003262586,0.003742988,0.01703308],"study_design_scores_gemma":[0.0004110531,0.00002224839,0.1063641,0.00002641747,0.0009013839,0.00001279056,0.008741886,0.863403,0.01914029,0.0002163344,0.0001665228,0.0005940325],"study_design_candidate":"observational","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"methods","genre_scores_codex":[0.9625749,0.0002456547,0.02305474,0.0002387896,0.00007323951,0.0003803305,0.01044642,0.00007599855,0.002909953],"genre_scores_gemma":[0.1454425,0.0003247687,0.8519383,0.0003580306,0.00001393916,0.00001118645,0.001614658,0.00001186222,0.0002847765],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.8541575,"threshold_uncertainty_score":0.9999951,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.03752216800121069,"score_gpt":0.2471145885669685,"score_spread":0.2095924205657578,"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."}}