Graph-RPI: predicting RNA–protein interactions via graph autoencoder and self-supervised learning strategies

Abstract RNA–protein interactions (RPIs) are essential for many biological functions and are associated with various diseases. Traditional methods for detecting RPIs are labor-intensive and costly, necessitating efficient computational methods. In this study, we proposed a novel sequence-based RPI p...

Full description

Saved in:

Bibliographic Details
Published in:	Briefings in bioinformatics Vol. 26; no. 3
Main Authors:	Guan, Jiahui, Yao, Lantian, Xie, Peilin, Zhao, Zhihao, Meng, Dian, Lee, Tzong-Yi, Wang, Junwen, Chiang, Ying-Chih
Format:	Journal Article
Language:	English
Published:	England Oxford University Press 01.05.2025 Oxford Publishing Limited (England)
Subjects:	Accuracy Algorithms Autoencoder Computational Biology - methods Graph neural networks Graphical representations Humans Machine learning Neural networks Neural Networks, Computer Predictions Problem Solving Protocol Protein interaction Proteins RNA - chemistry RNA - metabolism RNA-Binding Proteins - metabolism Self-supervised learning Supervised Machine Learning graph neural networks self-supervised learning multi-feature fusion RNA–protein interactions
ISSN:	1467-5463, 1477-4054, 1477-4054
Online Access:	Get full text
Tags:	Add Tag No Tags, Be the first to tag this record!

Description
Summary:	Abstract RNA–protein interactions (RPIs) are essential for many biological functions and are associated with various diseases. Traditional methods for detecting RPIs are labor-intensive and costly, necessitating efficient computational methods. In this study, we proposed a novel sequence-based RPI prediction framework based on graph neural networks (GNNs) that addressed key limitations of existing methods, such as inadequate feature integration and negative sample construction. Our method represented RNAs and proteins as nodes in a unified interaction graph, enhancing the representation of RPI pairs through multi-feature fusion and employing self-supervised learning strategies for model training. The model’s performance was validated through five-fold cross-validation, achieving accuracy of 0.880, 0.811, 0.950, 0.979, 0.910, and 0.924 on the RPI488, RPI369, RPI2241, RPI1807, RPI1446, and RPImerged datasets, respectively. Additionally, in cross-species generalization tests, our method outperformed existing methods, achieving an overall accuracy of 0.989 across 10 093 RPI pairs. Compared with other state-of-the-art RPI prediction methods, our approach demonstrates greater robustness and stability in RPI prediction, highlighting its potential for broad biological applications and large-scale RPI analysis.
Bibliography:	ObjectType-Article-1 SourceType-Scholarly Journals-1 ObjectType-Feature-2 content type line 14 content type line 23 Jiahui Guan, Lantian Yao and Peilin Xie contributed equally to this work.
ISSN:	1467-5463 1477-4054 1477-4054
DOI:	10.1093/bib/bbaf292