Cross‐Species Prediction of Transcription Factor Binding by Adversarial Training of a Novel Nucleotide‐Level Deep Neural Network
Advanced Science · 2024
DOI: 10.1002/advs.202405685 ↗Kurz gefasst
Fachartikel in der Zeitschrift Advanced Science, 2024. Verfasst von Qinhu Zhang und 4 weiteren Autor:innen. Beteiligte Einrichtung: Ningbo Institute of Digital Twin, Eastern Institute of Technology, Ningbo, 315201, China. Hauptthemen laut MEDLINE: Neural Networks, Computer, Transcription Factors, Polymorphism, Single Nucleotide. Die Arbeit wurde bisher 11-mal zitiert (Crossref) und nennt 55 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: STI 2030-Major Projects, National Natural Science Foundation of China, China Postdoctoral Science Foundation.
Abstract (Originalsprache)
Cross‐species prediction of TF binding remains a major challenge due to the rapid evolutionary turnover of individual TF binding sites, resulting in cross‐species predictive performance being consistently worse than within‐species performance. In this study, a novel Nucleotide‐Level Deep Neural Network (NLDNN) is first proposed to predict TF binding within or across species. NLDNN regards the task of TF binding prediction as a nucleotide‐level regression task, which takes DNA sequences as input and directly predicts experimental coverage values. Beyond predictive performance, it also assesses model performance by locating potential TF binding regions, discriminating TF‐specific single‐nucleotide polymorphisms (SNPs), and identifying causal disease‐associated SNPs. The experimental results show that NLDNN outperforms the competing methods in these tasks. Then, a dual‐path framework is designed for adversarial training of NLDNN to further improve the cross‐species prediction performance by pulling the domain space of human and mouse species closer. Through comparison and analysis, it finds that adversarial training not only can improve the cross‐species prediction performance between humans and mice but also enhance the ability to locate TF binding regions and discriminate TF‐specific SNPs. By visualizing the predictions, it is figured out that the framework corrects some mispredictions by amplifying the coverage values of incorrectly predicted peaks.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Stichwörter der Autor:innen
Substanzen
Autor:innen und Institutionen
- Ningbo Institute of Digital Twin, Eastern Institute of Technology, Ningbo, 315201, China.
- Division of Life Sciences and Medicine, University of Science and Technology of China, Hefei, 230021, China.
- Big Data and Intelligent Computing Research Center, Guangxi Academy of Science, Nanning, 530007, China.
- Institute for Regenerative Medicine, Shanghai East Hospital, Tongji University, Shanghai, 200092, China.
ORCID-Profile: Qinhu Zhang
Förderung
- STI 2030-Major Projects
- National Natural Science Foundation of China
- China Postdoctoral Science Foundation
- Natural Science Foundation of Ningbo City
- Key Research and Development Program of Ningbo City
- Key Project of Science and Technology of Guangxi
- Guangxi Natural Science Foundation
- Guangxi Science and Technology Base and Talents Special Project
Zitieren
APA
Zhang, Q., Wang, S., Li, Z., Pan, Y., & Huang, D. S. (2024). Cross‐Species Prediction of Transcription Factor Binding by Adversarial Training of a Novel Nucleotide‐Level Deep Neural Network. Advanced Science. https://doi.org/10.1002/advs.202405685
Vancouver
Zhang Q, Wang S, Li Z, Pan Y, Huang DS. Cross‐Species Prediction of Transcription Factor Binding by Adversarial Training of a Novel Nucleotide‐Level Deep Neural Network. Advanced Science. 2024. doi:10.1002/advs.202405685
BibTeX
@article{Zhang2024,
author = {Zhang, Qinhu and Wang, Siguo and Li, Zhipeng and Pan, Yijie and Huang, De-Shuang},
title = {Cross‐Species Prediction of Transcription Factor Binding by Adversarial Training of a Novel Nucleotide‐Level Deep Neural Network},
journal = {Advanced Science},
year = {2024},
doi = {10.1002/advs.202405685}
}
EUVIMED