{"id":"W3042732674","doi":"10.1101/2020.07.14.202960","title":"MetaWorks: A flexible, scalable bioinformatic pipeline for high-throughput multi-marker biodiversity assessments","year":2020,"lang":"en","type":"preprint","venue":"bioRxiv (Cold Spring Harbor Laboratory)","topic":"Environmental DNA in Biodiversity Studies","field":"Environmental Science","cited_by":9,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Guelph","funders":"","keywords":"Workflow; Scalability; Computer science; Pipeline (software); Computational biology; Biology; Data science; Database","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":[0.00433684,0.003889984,0.001586415,0.003094792,0.001488812,0.003620622,0.005351247,0.001779985,0.02301426],"category_scores_gemma":[0.005401992,0.002914715,0.003236402,0.002098225,0.0008603533,0.003440603,0.00417152,0.003973306,0.01816162],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001414195,"about_ca_system_score_gemma":0.002840632,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.002967235,"about_ca_topic_score_gemma":0.003193594,"domain_scores_codex":[0.9977716,0.0002754126,0.0001941151,0.0008118197,0.0007268774,0.0002201933],"domain_scores_gemma":[0.9975749,0.0007224205,0.0003232943,0.0005590473,0.0004850151,0.0003353951],"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":[0.00470644,0.0004551715,0.006769125,0.002810087,0.00151444,0.001168506,0.001058412,0.02450491,0.1916187,0.008687982,0.488649,0.268057],"study_design_scores_gemma":[0.002120179,0.0005617903,0.01236625,0.0005913039,0.0006323053,0.0009657434,0.0003498867,0.3980738,0.2315242,0.03964623,0.3120071,0.001161203],"study_design_candidate":"not_applicable","study_design_consensus":"not_applicable","genre_codex":"software","genre_gemma":"software","genre_scores_codex":[0.007014321,0.0003993856,0.4095404,0.0004399693,0.0002670882,0.0005956509,0.02630152,0.5521287,0.003312871],"genre_scores_gemma":[0.06514025,0.0006340243,0.7310919,0.001226661,0.0002264472,0.003169562,0.102533,0.08725312,0.008724992],"genre_candidate":"software","genre_consensus":"software","teacher_disagreement_score":0.02301426,"threshold_uncertainty_score":0.07699031,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02831425381916517,"score_gpt":0.2370567693451829,"score_spread":0.2087425155260178,"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."}}