{"id":"W4304783041","doi":"10.3389/fmicb.2022.1018237","title":"Development of a data science CURE in microbiology using publicly available microbiome datasets","year":2022,"lang":"en","type":"article","venue":"Frontiers in Microbiology","topic":"Genetics, Bioinformatics, and Biomedical Research","field":"Biochemistry, Genetics and Molecular Biology","cited_by":4,"is_retracted":false,"has_abstract":true,"ca_institutions":"Genome British Columbia; Canada's Michael Smith Genome Sciences Centre; University of British Columbia","funders":"University of British Columbia","keywords":"Undergraduate research; Big data; Data science; Microbiome; Computer science; Dissemination; Medical education; Biology; Bioinformatics; Medicine","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":[0.01989082,0.001580567,0.0008323452,0.002074022,0.001382151,0.004249035,0.0038031,0.002304294,0.007123971],"category_scores_gemma":[0.05415013,0.001191014,0.002087369,0.001145818,0.001762419,0.00575068,0.008862743,0.006646954,0.006463269],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00171959,"about_ca_system_score_gemma":0.005247616,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.002726213,"about_ca_topic_score_gemma":0.003738948,"domain_scores_codex":[0.9905565,0.003552348,0.0006147971,0.001723257,0.002773709,0.000779494],"domain_scores_gemma":[0.9786637,0.008076898,0.001219154,0.004533469,0.004714233,0.00279256],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"not_applicable","study_design_scores_codex":[0.001365066,0.00680507,0.03370886,0.002386478,0.0003077817,0.001067718,0.008177466,0.03913115,0.0473515,0.01906548,0.1037796,0.7368538],"study_design_scores_gemma":[0.001268069,0.004810837,0.01724431,0.001192905,0.0002206517,0.001128065,0.004088347,0.2722892,0.1065363,0.04659066,0.5439942,0.0006364926],"study_design_candidate":"not_applicable","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.1405927,0.0005734551,0.7189149,0.009876947,0.001203887,0.008773034,0.00784888,0.09228602,0.01993016],"genre_scores_gemma":[0.0724806,0.0002739246,0.9053033,0.001663013,0.00007828875,0.002338881,0.009421364,0.002303739,0.006136929],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.01989082,"threshold_uncertainty_score":0.1051939,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.03681867717887056,"score_gpt":0.2968595048081784,"score_spread":0.2600408276293078,"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."}}