{"id":"W2384183100","doi":"10.1101/053397","title":"RefSoil: A reference database of soil microbial genomes","year":2016,"lang":"en","type":"preprint","venue":"bioRxiv (Cold Spring Harbor Laboratory)","topic":"Genomics and Phylogenetic Studies","field":"Biochemistry, Genetics and Molecular Biology","cited_by":8,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of British Columbia","funders":"U.S. Department of Energy; Biological and Environmental Research; Office of Science; National Science Foundation","keywords":"Metagenomics; Phylum; Genome; Genomics; Reference database; Database; Biology; Archaea; Amplicon; Computational biology; Data science; Computer science; Gene; Genetics","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":[0.00296373,0.002022198,0.00167266,0.008546173,0.001055359,0.002035224,0.002688472,0.001631553,0.009378036],"category_scores_gemma":[0.007365612,0.0008997619,0.00107938,0.01189889,0.0004046163,0.001814382,0.001875814,0.001407834,0.01087022],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0008784215,"about_ca_system_score_gemma":0.003036943,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00471477,"about_ca_topic_score_gemma":0.005830884,"domain_scores_codex":[0.9980617,0.0003060763,0.0003047095,0.0006007016,0.0004962705,0.0002304909],"domain_scores_gemma":[0.9969904,0.0006935336,0.0004990057,0.000735804,0.0008124865,0.0002686583],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"bench_or_experimental","study_design_gemma":"not_applicable","study_design_scores_codex":[0.007231058,0.0005741481,0.0286538,0.01468094,0.0008348661,0.001375628,0.001065032,0.01469168,0.3589914,0.01552417,0.3360034,0.2203739],"study_design_scores_gemma":[0.0008013801,0.0006639178,0.03902423,0.001391328,0.0005844572,0.001517916,0.0004361368,0.01351501,0.06710538,0.007575822,0.86711,0.0002744128],"study_design_candidate":"not_applicable","study_design_consensus":null,"genre_codex":"dataset","genre_gemma":"dataset","genre_scores_codex":[0.04561813,0.004322481,0.04337598,0.0002935964,0.0001747772,0.000267708,0.8776183,0.02263234,0.005696636],"genre_scores_gemma":[0.01478945,0.0007566136,0.04234707,0.00006759843,0.00001952383,0.0001821822,0.9402046,0.001000239,0.0006327917],"genre_candidate":"dataset","genre_consensus":"dataset","teacher_disagreement_score":0.009378036,"threshold_uncertainty_score":0.03137267,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01655692830335351,"score_gpt":0.2219885538192902,"score_spread":0.2054316255159367,"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."}}