HetGAT-LMI:用于预测lncRNA-miRNA相互作用的可解释异构图注意方法
Ran Liu1, Zihao Wang1, Tianming Han1,2
1School of Computer Science and Software Engineering, University of Science and Technology Liaoning, Anshan, 114051, China.
Journal of chemical information and modeling
|December 23, 2025
概括
这项研究介绍了HetGAT-LMI,这是一种用于预测长非编码RNA (lncRNA) 和microRNA (miRNA) 相互作用的新型模型. 它提高了准确性和可解释性,为疾病机制研究提供了更好的工具.
科学领域:
- 计算生物学 计算生物学
- 生物信息学是一种生物信息学.
- 基因组学就是基因组学.
背景情况:
- 识别长非编码RNA (lncRNA) 和微RNA (miRNA) 相互作用对于理解基因调节和疾病至关重要.
- 目前的图形神经网络 (GNN) 方法在表示能力,边缘偏差和多模式特征融合方面存在局限性.
研究的目的:
- 开发一个先进的异质图注意力模型, HetGAT-LMI,用于准确预测 lncRNA-miRNA 相互作用.
- 通过整合不同的序列和结构特征来克服现有的GNN的局限性.
主要方法:
- 构建了一个异质的lncRNA-miRNA网络,具有多层次的相似性和相互作用边缘.
- 将多模式RNA特征 (K-mer,G-gap,CTD,RNA折叠结构) 融为统一的表示形式.
- 采用GATv2多头注意力和双向门融合,以实现强大的编码和歧视.
主要成果:
- HetGAT-LMI实现了高性能,AUC为0.9585和AUPR为0.9467. 这两种方法的AUC值为0.9467.
- SHAP分析显示了关键特征的重要性 (miRNA的CTD,lncRNA的MFE).
- 对HOTAIR和MALAT1的案例研究验证了该模型的生物相关性和外部有效性.
结论:
- HetGAT-LMI 在预测 lncRNA-miRNA 相互作用方面提供了更高的准确性,稳定性和可解释性.
- 该模型作为一种有价值的工具,用于高吞吐量选和产生关于相互作用机制的假设.
相关概念视频
lncRNA - Long Non-coding RNAs
9.7K
In humans, more than 80% of the genome gets transcribed. However, only around 2% of the genome codes for proteins. The remaining part produces non-coding RNAs which includes ribosomal RNAs, transfer RNAs, telomerase RNAs, and regulatory RNAs, among other types. A large number of regulatory non-coding RNAs have been classified into two groups depending upon their length – small non-coding RNAs, such as microRNA, which are less than 200 nucleotides in length, and long non-coding RNA...
9.7K
lncRNA - Long Non-coding RNAs
3.4K
3.4K
MicroRNAs
3.7K
MicroRNA (miRNA) are short, regulatory RNA transcribed from introns (non-coding regions of a gene) or intergenic regions (stretches of DNA present between genes). Several processing steps are required to form biologically active, mature miRNA. The initial transcript, called primary miRNA (pri-mRNA), base-pairs with itself, forming a stem-loop structure. Within the nucleus, an endonuclease enzyme, called Drosha, shortens the stem-loop structure into hairpin-shaped pre-miRNA. After the pre-miRNA...
3.7K


