{"id":"W2127325366","doi":"10.1186/1471-2105-4-65","title":"Statistical extraction of Drosophila cis-regulatory modules using exhaustive assessment of local word frequency","year":2003,"lang":"en","type":"article","venue":"BMC Bioinformatics","topic":"Genomics and Chromatin Dynamics","field":"Biochemistry, Genetics and Molecular Biology","cited_by":26,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"National Institute of General Medical Sciences; National Institutes of Health; York University","keywords":"Genome; DNA microarray; Coding (social sciences); Computational biology; Gene; Locus (genetics); Biology; Coding region; Regulatory sequence; Cis-regulatory module; Genetics; Computer science; Regulation of gene expression; Gene expression; Statistics; Mathematics; Enhancer","routes":{"ca_aff":false,"ca_fund":true,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.002126098,0.0006589715,0.0008739417,0.003597092,0.0004559598,0.000692779,0.0007191714,0.0008451166,0.0009414837],"category_scores_gemma":[0.01408977,0.0003101607,0.000887544,0.001551627,0.0004498083,0.001143895,0.0008617743,0.0004943002,0.0005442343],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.000475401,"about_ca_system_score_gemma":0.0008432693,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.001897666,"about_ca_topic_score_gemma":0.002888824,"domain_scores_codex":[0.9991311,0.00030356,0.0001071253,0.0002227387,0.0001719885,0.00006345571],"domain_scores_gemma":[0.9867665,0.01091031,0.0006652715,0.0006247943,0.0008673661,0.0001657125],"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":[0.001357715,0.0004407187,0.1005443,0.0008789396,0.0007161754,0.0005904516,0.0007146322,0.2219506,0.09065416,0.003159925,0.002744013,0.5762483],"study_design_scores_gemma":[0.000049861,0.0001540174,0.02093421,0.00002795162,0.00008786838,0.0003080985,0.0000911078,0.9495386,0.02299501,0.005078289,0.0006908444,0.00004413735],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"methods","genre_scores_codex":[0.5727126,0.0003900875,0.4227223,0.00009969526,0.00001234618,0.0001206943,0.001219919,0.002282746,0.0004396958],"genre_scores_gemma":[0.7633343,0.000148028,0.230215,0.0000690134,0.00002824253,0.0003049488,0.005235306,0.0002083146,0.0004567539],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.003597092,"threshold_uncertainty_score":0.011244,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01464972971313742,"score_gpt":0.2802310668427063,"score_spread":0.2655813371295688,"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."}}