{"id":"W4416003795","doi":"10.1038/s42003-025-08909-9","title":"Cross-platform motif discovery and benchmarking to explore binding specificities of poorly studied human transcription factors","year":2025,"lang":"en","type":"article","venue":"Communications Biology","topic":"Genomics and Chromatin Dynamics","field":"Biochemistry, Genetics and Molecular Biology","cited_by":5,"is_retracted":false,"has_abstract":true,"ca_institutions":"Occupational Cancer Research Centre; McGill University and Génome Québec Innovation Centre; BC Children's Hospital; University of British Columbia; Amgen (Canada)","funders":"National Institute of Arthritis and Musculoskeletal and Skin Diseases; National Institute of Allergy and Infectious Diseases; National Human Genome Research Institute; Natural Sciences and Engineering Research Council of Canada; National Science Foundation; Canadian Institutes of Health Research; Alliance de recherche numérique du Canada; Russian Science Foundation; National Institutes of Health; Schweizerischer Nationalfonds zur Förderung der Wissenschaftlichen Forschung; Canada Research Chairs; National Cancer Institute; University of Toronto; École Polytechnique Fédérale de Lausanne; Deutsche Forschungsgemeinschaft; Vetenskapsrådet","keywords":"Motif (music); Benchmarking; Sequence motif; Transcription factor; Codebook; Structural motif","routes":{"ca_aff":true,"ca_fund":true,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":false},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.00995668,0.001406705,0.001463421,0.002720269,0.0006627653,0.0009288233,0.001288328,0.0008380029,0.002284741],"category_scores_gemma":[0.01655328,0.0004551406,0.001453206,0.001961574,0.0006257903,0.001256923,0.002191062,0.0009018892,0.0009638682],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0004528066,"about_ca_system_score_gemma":0.0008132701,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.001356176,"about_ca_topic_score_gemma":0.002423642,"domain_scores_codex":[0.9955733,0.001590094,0.0003685903,0.00144059,0.0007086311,0.0003187438],"domain_scores_gemma":[0.9923493,0.00506083,0.0003766755,0.001307577,0.0006374268,0.0002682717],"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":[0.007173265,0.001426235,0.099819,0.004806272,0.003225309,0.0009003453,0.00100205,0.1744382,0.4610485,0.004131519,0.01174872,0.2302807],"study_design_scores_gemma":[0.0003280552,0.002260689,0.0430228,0.000127228,0.0005779195,0.0008980336,0.000337394,0.7096391,0.2265987,0.006375539,0.009601909,0.0002326446],"study_design_candidate":"bench_or_experimental","study_design_consensus":"bench_or_experimental","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.8608912,0.00181353,0.1183033,0.0001124255,0.0000788452,0.0002388406,0.006557623,0.01067549,0.00132885],"genre_scores_gemma":[0.7955768,0.0003863615,0.1774607,0.0001371706,0.00003084928,0.0006355275,0.02295722,0.00197305,0.0008423251],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.00995668,"threshold_uncertainty_score":0.05265653,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.06360928782051116,"score_gpt":0.3389688772160838,"score_spread":0.2753595893955726,"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."}}