{"id":"W4407251531","doi":"10.1038/s41597-025-04548-1","title":"An Integrated Database for Exploring Alternative Promoters in Animals","year":2025,"lang":"en","type":"article","venue":"Scientific Data","topic":"Genomics and Chromatin Dynamics","field":"Biochemistry, Genetics and Molecular Biology","cited_by":1,"is_retracted":false,"has_abstract":true,"ca_institutions":"Ministry of Agriculture","funders":"Huazhong Agricultural University; Fundamental Research Funds for the Central Universities; Natural Science Foundation of Hubei Province; National Natural Science Foundation of China","keywords":"Biology; Promoter; Enhancer; Computational biology; Database; Transcription factor; Transcription (linguistics); Genetics; Gene expression; Gene; Computer science","routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":false},"retraction":null,"screen":null,"direct_labels":[{"model":"gemma","categories":[],"domain":null,"study_design":"not_applicable","genre":"dataset","about_ca_system":false,"about_ca_topic":false,"confidence":"low","status":"direct model label, unvalidated"},{"model":"gpt","categories":[],"domain":null,"study_design":"not_applicable","genre":"dataset","about_ca_system":false,"about_ca_topic":false,"confidence":"low","status":"direct model label, unvalidated"}],"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.00198731,0.002772472,0.002396934,0.009383588,0.0007550389,0.002428577,0.002976801,0.002234721,0.02718659],"category_scores_gemma":[0.004688854,0.001365221,0.001467131,0.01028873,0.0005343966,0.003025044,0.003113205,0.00195297,0.02331063],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0008848736,"about_ca_system_score_gemma":0.002038309,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.002351938,"about_ca_topic_score_gemma":0.00268089,"domain_scores_codex":[0.9990416,0.0001440047,0.0002658881,0.0002533465,0.0002073674,0.00008777624],"domain_scores_gemma":[0.9980245,0.0007893603,0.000338584,0.0003491929,0.0002371525,0.0002613023],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"not_applicable","study_design_gemma":"not_applicable","study_design_scores_codex":[0.003597513,0.0008269957,0.01639049,0.03700231,0.001326471,0.003611945,0.001944587,0.007426452,0.1284842,0.01975922,0.5021179,0.2775119],"study_design_scores_gemma":[0.0004255177,0.000249814,0.01363663,0.001467098,0.0004869883,0.00157562,0.0003468753,0.006494602,0.01368806,0.009808585,0.951597,0.0002232126],"study_design_candidate":"not_applicable","study_design_consensus":"not_applicable","genre_codex":"dataset","genre_gemma":"dataset","genre_scores_codex":[0.006734607,0.00473575,0.03533151,0.0002234337,0.0001236849,0.0003014313,0.8971324,0.0472367,0.008180378],"genre_scores_gemma":[0.007921058,0.002162251,0.03927432,0.0001627606,0.0000370629,0.0004116518,0.9458967,0.002892802,0.001241512],"genre_candidate":"dataset","genre_consensus":"dataset","teacher_disagreement_score":0.02718659,"threshold_uncertainty_score":0.09094822,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.06597041708514718,"score_gpt":0.3276484842748255,"score_spread":0.2616780671896783,"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."}}