{"id":"W3038449881","doi":"10.1093/nargab/lqaa046","title":"On the prediction of DNA-binding preferences of C2H2-ZF domains using structural models: application on human CTCF","year":2020,"lang":"en","type":"article","venue":"NAR Genomics and Bioinformatics","topic":"RNA and protein synthesis mechanisms","field":"Biochemistry, Genetics and Molecular Biology","cited_by":14,"is_retracted":false,"has_abstract":true,"ca_institutions":"BC Children's Hospital; University of British Columbia","funders":"European Regional Development Fund; Ministerio de Ciencia e Innovación","keywords":"Zinc finger; CTCF; DNA; Transcription factor; Computational biology; Binding site; Biology; DNA-binding protein; Chromatin; Genetics; Gene","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":[0.0001197878,0.00009659462,0.0001149264,0.00002416112,0.0001036971,0.00001466738,0.0001256864,0.00009457122,0.000002879774],"category_scores_gemma":[0.00001725234,0.00006691779,0.00004197163,0.00004401351,0.00005012449,0.000006859525,0.00004869004,0.0000494138,6.008881e-7],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.000008625849,"about_ca_system_score_gemma":0.00002237644,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.000004544407,"about_ca_topic_score_gemma":0.00000119085,"domain_scores_codex":[0.9994147,0.00001899854,0.0002734354,0.00009843292,0.0001078213,0.00008664189],"domain_scores_gemma":[0.999529,0.00001528464,0.000231873,0.000154125,0.00003532717,0.00003436852],"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":[0.000055377,0.000008398561,0.00004368558,0.00006542575,0.00003043671,2.985455e-8,0.0005715993,0.0009022821,0.9890482,0.005876251,0.00001891926,0.003379432],"study_design_scores_gemma":[0.0002443378,0.0006405478,0.00005441788,0.00003117851,0.00002557786,0.000001852384,0.0006623276,0.1018816,0.892814,0.003388271,0.0001531374,0.0001026885],"study_design_candidate":"bench_or_experimental","study_design_consensus":"bench_or_experimental","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9917113,0.00003186824,0.007353695,0.00005093926,0.0000242948,0.0002572371,0.0001060549,0.000004005415,0.0004606448],"genre_scores_gemma":[0.9968095,0.00007794092,0.002917106,0.00008256146,0.00004825135,0.000004639796,0.00004920793,0.000007111232,0.00000368129],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.1009794,"threshold_uncertainty_score":0.2728828,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.04841538685684868,"score_gpt":0.2496151709710039,"score_spread":0.2011997841141553,"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."}}