{"id":"W2769058621","doi":"10.1093/nar/gkx1145","title":"Expanding the repertoire of DNA shape features for genome-scale studies of transcription factor binding","year":2017,"lang":"en","type":"article","venue":"Nucleic Acids Research","topic":"Genomics and Chromatin Dynamics","field":"Biochemistry, Genetics and Molecular Biology","cited_by":114,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"National Institute of General Medical Sciences; National Human Genome Research Institute; Banff International Research Station for Mathematical Innovation and Discovery","keywords":"Bioconductor; Biology; Interpretability; Computational biology; DNA; DNA binding site; DNA sequencing; Base pair; Transcription factor; Genetics; Biological system; Gene; Computer science; Artificial intelligence; Promoter; Gene expression","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":[0.000683876,0.00009765993,0.0001859112,0.00005493056,0.0006003384,0.00004407172,0.0005756317,0.0001183007,0.00001092705],"category_scores_gemma":[0.0003382638,0.00006998307,0.0001258145,0.0000431773,0.0004383642,0.000006366737,0.0002091882,0.0001182179,8.530242e-7],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00002575865,"about_ca_system_score_gemma":0.00004919045,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00001400253,"about_ca_topic_score_gemma":0.00004235464,"domain_scores_codex":[0.9989907,0.00005927782,0.0002216647,0.0002342576,0.0002340004,0.0002601013],"domain_scores_gemma":[0.9988735,0.0000510806,0.0001420668,0.0006014099,0.0002956266,0.00003631682],"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":[0.00008667284,0.00002705498,0.00128374,0.000140453,0.00008784669,2.726496e-7,0.0009395128,0.000005902446,0.9937377,0.00006215059,0.000247621,0.003381049],"study_design_scores_gemma":[0.0007403059,0.0006359412,0.03086166,0.00007974054,0.00002242212,0.000004306039,0.004941032,0.0003305714,0.9592363,0.000351576,0.002634223,0.0001619255],"study_design_candidate":"bench_or_experimental","study_design_consensus":"bench_or_experimental","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.997687,0.001023413,0.00008713997,0.0003234454,0.00008541835,0.0003295649,0.0001086748,0.000002973464,0.0003523244],"genre_scores_gemma":[0.997571,0.0008910207,0.000629729,0.000009162306,0.0001160884,0.00003080782,0.00002713993,0.00002160289,0.0007034802],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.03450143,"threshold_uncertainty_score":0.4617379,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.06877315354525615,"score_gpt":0.3671946632372093,"score_spread":0.2984215096919531,"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."}}