{"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":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0002171967,0.0001465724,0.0002102428,0.00005429682,0.00004346056,0.00001002652,0.0001099679,0.0001598045,0.00001965111],"category_scores_gemma":[0.00003717337,0.0001445555,0.00007666244,0.00007355887,0.0001867852,0.000009736208,0.00004371177,0.00007919884,0.000001143213],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00004574403,"about_ca_system_score_gemma":0.0002446141,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00001038388,"about_ca_topic_score_gemma":0.00001109555,"domain_scores_codex":[0.9989154,0.00003084441,0.0005701833,0.0001211823,0.0001878907,0.0001744803],"domain_scores_gemma":[0.999117,0.00002140675,0.0003556158,0.0003060493,0.0001368856,0.00006302699],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"bench_or_experimental","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.0001266655,0.0008414956,0.04414722,0.002188632,0.0003939297,0.000004148707,0.0004320395,0.1349496,0.6953043,0.1147285,0.0002921564,0.006591346],"study_design_scores_gemma":[0.001385752,0.0005840884,0.0733507,0.0001062852,0.0001423754,0.00007540022,0.002418913,0.8279811,0.08957,0.003615513,0.0001872906,0.0005825498],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.4895257,0.00004578364,0.5093654,9.499715e-7,0.00007028602,0.00009119197,0.00006060937,0.000002835743,0.0008372347],"genre_scores_gemma":[0.6399721,0.00002106715,0.3598925,0.000008143287,0.0000146023,0.000002259449,0.00006927741,0.000009872469,0.00001012855],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.6930315,"threshold_uncertainty_score":0.5894802,"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."}}