{"id":"W2821433317","doi":"10.7717/peerj.6304","title":"Prioritizing bona fide bacterial small RNAs with machine learning classifiers","year":2019,"lang":"en","type":"article","venue":"PeerJ","topic":"Genomics and Phylogenetic Studies","field":"Biochemistry, Genetics and Molecular Biology","cited_by":10,"is_retracted":false,"has_abstract":true,"ca_institutions":"Memorial University of Newfoundland","funders":"Natural Sciences and Engineering Research Council of Canada","keywords":"Biology; Computational biology; Genome; Context (archaeology); Precision and recall; Artificial intelligence; ORFS; Pipeline (software); Open reading frame; Machine learning; Bacterial genome size; Genetics; Gene; Computer science; Peptide sequence","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":[0.002026764,0.001774406,0.001157265,0.001865755,0.0005254105,0.001413531,0.00107739,0.001370328,0.001205844],"category_scores_gemma":[0.004098294,0.0003674886,0.0009252478,0.0006452344,0.0003750238,0.001454426,0.0007493111,0.001503208,0.0009713216],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0007757879,"about_ca_system_score_gemma":0.001149578,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00196385,"about_ca_topic_score_gemma":0.00199277,"domain_scores_codex":[0.9990048,0.0002653077,0.00006575925,0.000265555,0.0002766148,0.0001220284],"domain_scores_gemma":[0.9979754,0.001163627,0.0002019107,0.0001064917,0.0004379762,0.0001147058],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.0006824491,0.0006468827,0.03021716,0.0005079155,0.00030901,0.0004275222,0.0001876318,0.3844822,0.06959741,0.002397724,0.006037478,0.5045066],"study_design_scores_gemma":[0.00001087524,0.00008354191,0.001155704,0.00001483754,0.00002511713,0.00003113452,0.00002595494,0.9882784,0.008004955,0.001542203,0.0008145951,0.00001257267],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.4057095,0.003053441,0.577872,0.0007852862,0.0003197321,0.0003084905,0.0009924037,0.006199839,0.004759188],"genre_scores_gemma":[0.7408561,0.0005719093,0.2537269,0.0003421811,0.0001544758,0.0002101649,0.002065508,0.000194307,0.001878439],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.002026764,"threshold_uncertainty_score":0.0107187,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.009201968950497007,"score_gpt":0.1981520699552252,"score_spread":0.1889501010047282,"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."}}