Deep learning the TF regulatory code for gene expression
Genome Research · 2026-06 · 36 (6) · 1238-1249
DOI: 10.1101/gr.281425.125 ↗Kurz gefasst
Fachartikel in der Zeitschrift Genome Research, online erschienen am 15.05.2026. Verfasst von Yuanyuan Li und 6 weiteren Autor:innen. Beteiligte Einrichtung: Yangtze Delta Region Institute (Quzhou), University of Electronic Science and Technology of China, Quzhou 324003, China. Hauptthemen laut MEDLINE: Transcription Factors, Deep Learning, Gene Expression Regulation, Regulatory Elements, Transcriptional. Die Arbeit nennt 41 Literaturangaben. Gefördert von: National Natural Science Foundation of China, Municipal Government of Quzhou.
Abstract (Originalsprache)
Gene transcription is activated through the interaction between cis -regulatory elements (CREs) and transcription factors (TFs). CREs serve as templates to provide binding sites, whereas TFs provide functions to directly initiate transcription. Current research mainly focuses on deciphering cis -regulatory code, while neglecting TF regulatory code. However, CREs alone are not sufficient to determine the binding of TFs, which makes the interpretation of cis -regulatory code ambiguous. In this study, we systematically analyze 13 TF binding profiles associated with transcription initiation and encode them as TF sequence to explore the TF regulatory code. Furthermore, we propose a deep learning model named DeepTF to predict gene expression from TF sequence. Results show that TF binding exhibits conserved positional preferences and combinatorial patterns in promoters and DeepTF is able to predict gene expression with high accuracy (AUROC = 0.97). Meanwhile, cross-cell-line validation (AUROC > 0.90) further confirms the model's transferability. Model interpretation reveals DeepTF successfully captures the TF regulatory grammar associated with gene expression. Compared with the cis -regulatory code, our proposed TF regulatory code is better suited for investigating the relationship between TF binding positions or combinatorial patterns and gene expression. Collectively, DeepTF simplifies gene expression prediction and provides clear biological insights of transcriptional regulation.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Substanzen
Autor:innen und Institutionen
- Yangtze Delta Region Institute (Quzhou), University of Electronic Science and Technology of China, Quzhou 324003, China.
- College of Computer Science and Technology, Xinjiang University, Urumqi 830017, China.
- School of Life Science and Technology and Center for Informational Biology, University of Electronic Science and Technology of China, Chengdu 610054, China.
- Computer Science and Information Systems, Murray State University, Murray, Kentucky 42071, USA.
- School of Life Science and Technology and Center for Informational Biology, University of Electronic Science and Technology of China, Chengdu 610054, China; hding@uestc.edu.cn zouquan@nclab.net liliu2010imu@163.com.
- Yangtze Delta Region Institute (Quzhou), University of Electronic Science and Technology of China, Quzhou 324003, China; hding@uestc.edu.cn zouquan@nclab.net liliu2010imu@163.com.
Förderung
- National Natural Science Foundation of China
- Municipal Government of Quzhou
Zitieren
APA
Li, Y., Huang, X., Qi, Y., Zhang, Z., Ding, H., Zou, Q., & Liu, L. (2026). Deep learning the TF regulatory code for gene expression. Genome Research, 36(6), 1238-1249. https://doi.org/10.1101/gr.281425.125
Vancouver
Li Y, Huang X, Qi Y, Zhang Z, Ding H, Zou Q, et al. Deep learning the TF regulatory code for gene expression. Genome Research. 2026;36(6):1238-1249. doi:10.1101/gr.281425.125
BibTeX
@article{Li2026,
author = {Li, Yuanyuan and Huang, Xiaohui and Qi, Yixuan and Zhang, Zheng and Ding, Hui and Zou, Quan and Liu, Li},
title = {Deep learning the TF regulatory code for gene expression},
journal = {Genome Research},
year = {2026},
volume = {36},
number = {6},
pages = {1238--1249},
doi = {10.1101/gr.281425.125}
}
EUVIMED