{"id":"W2168872202","doi":"10.1093/bioinformatics/bth196","title":"Identifying multiple alignment regions satisfying simple formulas and patterns","year":2004,"lang":"en","type":"article","venue":"Bioinformatics","topic":"Genomics and Phylogenetic Studies","field":"Biochemistry, Genetics and Molecular Biology","cited_by":9,"is_retracted":false,"has_abstract":true,"ca_institutions":"McGill University","funders":"National Institutes of Health","keywords":"Simple (philosophy); Documentation; Set (abstract data type); Computer science; Software; Data mining; Theoretical computer science; Algorithm; Programming language","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":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.00007699632,0.0001291913,0.0001074211,0.00003022096,0.0001673099,0.00004372174,0.00008956456,0.00006485874,0.000001841263],"category_scores_gemma":[0.00003003677,0.0001206368,0.00004850372,0.00003259906,0.00003825184,0.000002640866,0.0001877149,0.00003759054,0.000006192819],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0000172155,"about_ca_system_score_gemma":0.00002488671,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00004882735,"about_ca_topic_score_gemma":0.0001289883,"domain_scores_codex":[0.9993317,0.000005785345,0.0002341694,0.0001243119,0.00008977906,0.000214247],"domain_scores_gemma":[0.9995933,0.000008875522,0.00008227241,0.0002152599,0.0000295633,0.00007070673],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"bench_or_experimental","study_design_gemma":"bench_or_experimental","study_design_scores_codex":[0.0001116994,0.0002748942,0.3157621,0.001019572,0.0009455206,0.00001602594,0.01485243,0.003526599,0.6019577,0.008221234,0.003365435,0.04994675],"study_design_scores_gemma":[0.01260143,0.00168672,0.3505272,0.0003397038,0.0003053301,0.0003342282,0.01796193,0.00494916,0.4619283,0.01605613,0.13012,0.00318981],"study_design_candidate":"bench_or_experimental","study_design_consensus":"bench_or_experimental","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.965106,0.0004343509,0.03366771,0.00008507735,0.00009477593,0.0001834721,0.00004342189,0.000006580732,0.0003786168],"genre_scores_gemma":[0.9860878,0.0007401473,0.01274259,0.0002493834,0.00007019704,0.00001413718,0.00005121595,0.00001251271,0.00003200734],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.1400294,"threshold_uncertainty_score":0.4919424,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02293349849612335,"score_gpt":0.2500734592962721,"score_spread":0.2271399608001487,"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."}}