MambaCAttnGCN+:一个全面的框架,整合了MambaTextCNN,交叉注意力和图形卷积网络,用于piRNA-疾病关联预测
Dengju Yao1, Xiangkui Li2, Xiaojuan Zhan3
1School of Computer Science and Technology, Harbin University of Science and Technology, Harbin, 150080, China. ydkvictory@hrbust.edu.c.
Scientific reports
|July 11, 2025
概括
这项研究介绍了MambaCAttnGCN+,这是一种用于预测piRNA-疾病关联的新型计算模型. 它准确地识别了piwi相互作用RNAs (piRNAs) 和疾病之间的联系,有助于生物医学研究.
科学领域:
- 生物医学信息学是生物医学信息学.
- 计算生物学是一种计算生物学.
- 基因组学就是基因组学.
背景情况:
- 了解piwi相互作用RNA (piRNA) 和疾病相互作用对于诊断和治疗至关重要.
- 现有的计算方法在稀疏的数据上扎,限制了准确的piRNA-疾病关联预测.
研究的目的:
- 开发一种更准确的计算模型来预测piRNA与疾病的关联.
- 利用异质图形构造和先进的序列嵌入来改善特征提取.
主要方法:
- 构建了一个整合piRNA序列,疾病语义和已知的关联的异质图.
- 使用MambaTextCNN用于piRNA序列特征提取和异质图形卷曲与交叉注意.
- 利用积极的未标记学习来开发MambaCAttnGCN+预测模型.
主要成果:
- 在两个数据集的5倍交叉验证中,MambaCAttnGCN+实现了0.94和0.953的高AUC.
- 该模型在预测piRNA-疾病关联方面表现优于其他七种最先进的方法.
- 除研究证实了MambaTextCNN在提取序列节点特征方面的卓越性能.
结论:
- MambaCAttnGCN+显示出作为对piRNA-疾病关联的预测工具的巨大潜力.
- 这些发现强调了整合序列信息,图形网络和高级学习技术的有效性.
- 这种方法有助于研究piRNA在各种疾病中的作用.
相关概念视频
piRNA - Piwi-interacting RNAs
7.0K
PIWI-interacting RNAs, or piRNAs, are the most abundant short non-coding RNAs. More than 20,000 genes have been found in humans that code for piRNAs while only 2000 genes have been found for miRNAs. piRNAs can act at the transcriptional and post-transcriptional levels and have a vital role in silencing transposable elements present in germ cells. They are also involved in epigenetic silencing and activation. Previously, they were thought to function only in germ cells but new evidence suggests...
7.0K
Genome-wide Association Studies-GWAS
14.3K
Genome-wide association studies or GWAS are used to identify whether common SNPs are associated with certain diseases. Suppose specific SNPs are more frequently observed in individuals with a particular disease than those without the disease. In that case, those SNPs are said to be associated with the disease. Chi-square analysis is performed to check the probability of the allele likely to be associated with the disease.
GWAS does not require the identification of the target gene involved in...
GWAS does not require the identification of the target gene involved in...
14.3K


