Volume 31, No 5, May 2021

ISSN: 1001-0602 
EISSN: 1748-7838 2018 
impact factor 17.848* 
(Clarivate Analytics, 2019)

Volume 31 Issue 5, May 2021: 495-516   |  Open Access

ORIGINAL ARTICLES

Predicting dynamic cellular protein–RNA interactions by deep learning using in vivo RNA structures

Lei Sun^1,2 , Kui Xu^1,2 , Wenze Huang^1,2 , Yucheng T. Yang^3,4 , Pan Li^1,2 , Lei Tang^1,2 , Tuanlin Xiong^1,2 , Qiangfeng Cliff Zhang^1,2,*

¹MOE Key Laboratory of Bioinformatics, Beijing Advanced Innovation Center for Structural Biology and Frontier Research Center for Biological Structure, Center for Synthetic and Systems Biology, School of Life Sciences, Tsinghua University, Beijing 100084, China
²Tsinghua-Peking Center for Life Sciences, Beijing 100084, China
³Program in Computational Biology and Bioinformatics, Yale University, New Haven, CT 06520, USA
⁴Department of Molecular Biophysics and Biochemistry, Yale University, New Haven, CT 06520, USA
These authors contributed equally: Lei Sun, Kui Xu, Wenze Huang, Yucheng T. Yang Correspondence: Qiangfeng Cliff Zhang(qczhang@tsinghua.edu.cn)

Interactions with RNA-binding proteins (RBPs) are integral to RNA function and cellular regulation, and dynamically reflect specific cellular conditions. However, presently available tools for predicting RBP–RNA interactions employ RNA sequence and/or predicted RNA structures, and therefore do not capture their condition-dependent nature. Here, after profiling transcriptome-wide in vivo RNA secondary structures in seven cell types, we developed PrismNet, a deep learning tool that integrates experimental in vivo RNA structure data and RBP binding data for matched cells to accurately predict dynamic RBP binding in various cellular conditions. PrismNet results for 168 RBPs support its utility for both understanding CLIP-seq results and largely extending such interaction data to accurately analyze additional cell types. Further, PrismNet employs an “attention” strategy to computationally identify exact RBP-binding nucleotides, and we discovered enrichment among dynamic RBP-binding sites for structure-changing variants (riboSNitches), which can link genetic diseases with dysregulated RBP bindings. Our rich profiling data and deep learning-based prediction tool provide access to a previously inaccessible layer of cell-type-specific RBP–RNA interactions, with clear utility for understanding and treating human diseases.

https://doi.org/10.1038/s41422-021-00476-y

FULL TEXT | PDF

Browse 1060