LPI-MAM:以miRNAs为中介基于深度学习预测lncRNA-蛋白质相互作用  

LPI-MAM: Predicting lncRNA-Protein Interactions with miRNAs as Mediators Based on Deep Learning

作　　者：屈文燕 颜静 李晓毅[1] 谭建军[1] 

机构地区：[1]北京工业大学环境与生命学院生物医学工程系,北京

出　　处：《计算生物学》2023年第2期11-21,共11页Hans Journal of Computational Biology

摘　　要：长链非编码RNA (Long non-coding RNAs, lncRNAs)是细胞增殖和死亡的重要调控因子,它的失调可能会导致多种疾病发生。LncRNAs主要是通过与蛋白质相互作用(lncRNA-protein interactions, lncRPIs)来发挥生物学功能。因此,研究lncRPIs对了解lncRNAs的功能及相关疾病至关重要。目前,多数计算方法依赖于已知的验证过的lncRPIs构建模型,但经过实验验证的样本是有限的。MiRNAs主要是与mRNAs结合导致基因沉默,而lncRNAs可作为竞争性内源性RNA,竞争性的结合miRNAs来间接地调节基因表达。本文提出LPI-MAM方法,使用miRNAs作为中间体来扩大lncRPIs的预测范围。该方法将序列、结构和组成转换分布特征融合,输入卷积神经网络和独立循环神经网络的集成深度学习框架中。结果表明,LPI-MAM在基准数据集上取得了良好的性能。并且通过构建可视化交互网络发现该模型具有预测未知lncRPIs的能力。Long non-coding RNAs (lncRNAs) are crucial regulatory factors of cell proliferation and death, its dysregulation may lead to the occurrence of a variety of diseases. LncRNAs play biological functions mainly through lncRNA-protein interactions (lncRPIs). Therefore, it becomes essential to study the interactions between lncRNA and protein (lncRPIs) for exploring the function of lncRNAs. At pre-sent, almost computational methods depend on known lncRPIs to build a model. However, the samples that have been verified are limited. MiRNAs mainly bind to mRNAs to cause gene silencing. As competitive endogenous RNAs (ceRNAs), lncRNAs can indirectly regulate gene expression by competitively binding miRNAs. This study proposes the LPI-MAM method, which uses miRNAs as mediators to expand the prediction range of lncRPIs. The features of sequence, structure and composition transformation distribution (CTD) are fused and then input into the integrated deep learning framework of convolutional neural network (CNN) and independent recurrent neural network (IndRNN). The results indicate that LPI-MAM has achieved good performance on benchmark dataset. And by constructing a visual interaction network, it is found that the model has the ability to predict unknown lncRPIs.

关 键 词：卷积神经网络 循环神经网络 深度学习 蛋白质相互作用 特征融合 细胞增殖 可视化交互 MIRNAS 

分 类 号：R73[医药卫生—肿瘤]