{"id":"W3043455078","doi":"10.1101/2019.12.12.865014","title":"coil: an R package for cytochrome C oxidase I (COI) DNA barcode data cleaning, translation, and error evaluation","year":2019,"lang":"en","type":"preprint","venue":"bioRxiv (Cold Spring Harbor Laboratory)","topic":"Environmental DNA in Biodiversity Studies","field":"Environmental Science","cited_by":6,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Guelph","funders":"Government of Canada; Ontario Genomics; Compute Canada; Genome Canada","keywords":"Barcode; DNA barcoding; Indel; Computational biology; Biology; Sequence analysis; Reading frame; Computer science; Error detection and correction; Frame (networking); Translation (biology); Genetics; Open reading frame; DNA; Gene; Algorithm; Peptide sequence; 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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.01034764,0.004305318,0.002930985,0.00330865,0.001349738,0.003578886,0.003540921,0.001340711,0.04619326],"category_scores_gemma":[0.04575513,0.002348879,0.003207089,0.002190514,0.001495927,0.002787349,0.004037612,0.004220264,0.05413499],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001095843,"about_ca_system_score_gemma":0.004181581,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.002912668,"about_ca_topic_score_gemma":0.003402621,"domain_scores_codex":[0.9923692,0.00263317,0.000812706,0.001924587,0.001809706,0.0004506065],"domain_scores_gemma":[0.9821544,0.00934999,0.002168055,0.002865191,0.00300886,0.000453531],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"not_applicable","study_design_gemma":"not_applicable","study_design_scores_codex":[0.001008569,0.0001184582,0.008339087,0.00494109,0.001409679,0.000568495,0.0007903011,0.01068455,0.01628667,0.01003179,0.7966439,0.1491773],"study_design_scores_gemma":[0.0007084784,0.0004096183,0.01395151,0.001421007,0.001044346,0.0012914,0.0003488953,0.1067871,0.06756572,0.04627891,0.7594301,0.0007629496],"study_design_candidate":"not_applicable","study_design_consensus":"not_applicable","genre_codex":"methods","genre_gemma":"software","genre_scores_codex":[0.006048698,0.0008529868,0.4656198,0.0007346821,0.0005030957,0.0005719778,0.06778171,0.4540341,0.003852849],"genre_scores_gemma":[0.03569563,0.0007708948,0.6404076,0.001214452,0.0002723183,0.0033846,0.07530953,0.2364081,0.006536806],"genre_candidate":"software","genre_consensus":null,"teacher_disagreement_score":0.04619326,"threshold_uncertainty_score":0.1545319,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.06100834051129319,"score_gpt":0.2666376014966184,"score_spread":0.2056292609853252,"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."}}