SeqMG-RPI: A Sequence-Based Framework Integrating Multi-Scale RNA Features and Protein Graphs for RNA-Protein Interaction Prediction

作者：Teng Ma, Mingjian Jiang, Shunpeng Pang, Zhi Zhang, Huaibin Hang, Wei Zhou, Yuanyuan Zhang · 发表于：Journal of Chemical Information and Modeling · 年份：2025 · DOI：10.1021/acs.jcim.5c00176 · 被引用次数：3 · 研究领域：RNA and protein synthesis mechanisms、RNA Research and Splicing、RNA modifications and cancer

RNA-protein interaction (RPI) plays a crucial role in cell biology, and accurate prediction of RPI is essential to understand molecular mechanisms and advance disease research. Some existing RPI prediction methods typically rely on a single feature and there is significant room for improvement. In this paper, we propose a novel sequence-based RPI prediction method, called SeqMG-RPI. For RNA, SeqMG-RPI introduces an innovative multi-scale RNA feature that integrates three sequence-based representations: a multi-channel RNA feature, a k -mer frequency feature, and a k -mer sparse matrix feature. For protein, SeqMG-RPI utilizes a graph-based protein feature to capture protein information. Moreover, a novel neural network architecture is constructed for feature extraction and RPI prediction. Through experiments from multiple perspectives across various datasets, it is demonstrated that the proposed method outperforms existing methods, which has better performance and generalization.