{"id":"W6930693233","doi":"10.5281/zenodo.16616314","title":"BOLDistilled: Comprehensive but compact DNA barcode reference libraries","year":2025,"lang":"en","type":"dataset","venue":"Zenodo (CERN European Organization for Nuclear Research)","topic":"Environmental DNA in Biodiversity Studies","field":"Environmental Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Guelph","funders":"Government of Canada","keywords":"Barcode; Scripting language; Taxonomy (biology); Documentation; Open source; DNA barcoding","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":[0.003041695,0.002566415,0.002612521,0.0065719,0.002515205,0.004427533,0.006375717,0.00298112,0.2549887],"category_scores_gemma":[0.01599641,0.003668472,0.002203672,0.00638777,0.0007511309,0.003477979,0.003849108,0.005258469,0.2878664],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001786198,"about_ca_system_score_gemma":0.004476327,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.003789752,"about_ca_topic_score_gemma":0.009791769,"domain_scores_codex":[0.9964517,0.0005559311,0.0004089797,0.0009328807,0.00131567,0.0003347764],"domain_scores_gemma":[0.9943565,0.001795208,0.0007090507,0.00141856,0.001269551,0.0004511012],"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":[0.0006584101,0.0001124248,0.0009040047,0.002968502,0.00010616,0.0002205556,0.0002606528,0.001032316,0.03760361,0.005272656,0.9025594,0.04830135],"study_design_scores_gemma":[0.0001567339,0.00009547093,0.002038484,0.0005356029,0.00008377783,0.0003441628,0.0001035431,0.001579968,0.04266565,0.003791047,0.9484605,0.0001450694],"study_design_candidate":"not_applicable","study_design_consensus":"not_applicable","genre_codex":"dataset","genre_gemma":"dataset","genre_scores_codex":[0.002664536,0.001030652,0.1354035,0.0006292675,0.00102467,0.0009096449,0.7783274,0.06088593,0.01912427],"genre_scores_gemma":[0.004561289,0.0005897559,0.1454628,0.000659811,0.0001678905,0.001588624,0.8000619,0.02418841,0.0227195],"genre_candidate":"dataset","genre_consensus":"dataset","teacher_disagreement_score":0.2549887,"threshold_uncertainty_score":0.8530222,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.03816032070721818,"score_gpt":0.2350572215192705,"score_spread":0.1968969008120524,"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."}}