基于深度学习的长链非编码RNA与微小RNA相互作用预测的研究进展  

Research Progress of Predicting Long Non-Coding RNA-MicroRNA Interaction Based on Deep Learning

作　　者：王文雅 

机构地区：[1]辽宁科技大学计算机与软件工程学院,辽宁 鞍山

出　　处：《数据挖掘》2022年第2期152-160,共9页Hans Journal of Data Mining

摘　　要：长链非编码RNA (lncRNA)与微小RNA (miRNA)都是非编码RNA,越来越多的证据表明两者之间的相互作用与癌症的发展、基因调控、细胞代谢等生物学过程高度相关。与此同时,随着RNA序列技术的快速发展人们发现了许多新的lncRNA和miRNA,这可能有助于探索新的基因调控模式,人们对于lncRNA-miRNA相互作用的研究兴趣也随之越来越大。为此,我们回顾了目前lncRNA-miRNA相互作用关系预测的研究进展,我们针对部分研究人员的最新研究成果着重调查了他们使用的计算方法和数据库。调查结果显示深度学习已经成为lncRNA-miRNA相互作用关系预测的首选策略,这可能是由于深度学习基础设施和专业知识的快速增长。虽然这些方法中有许多都有明显的局限性,但深度学习有望在未来lncRNA-miRNA相互作用关系预测的领域取得更加充分的应用。Long non-coding RNA (LncRNA) and microRNA (miRNA) are both noncoding RNA. More and more evidence shows that the interaction between them is highly related to biological processes such as cancer development, gene regulation and cell metabolism. At the same time, with the rapid devel-opment of RNA sequence measuring technology, many new lncRNAs and miRNAs have been found, which may help to explore new gene regulation modes, and people are more and more interested in the research of lncRNA-miRNA interaction. Therefore, we reviewed the current research progress in the prediction of lncRNA-miRNA interaction. We focused on the calculation methods and databases used by some researchers according to their latest research results. The results show that deep learning has become the preferred strategy for the prediction of lncRNA-miRNA interaction, which may be due to the rapid growth of deep learning infrastructure and expertise. Although many of these methods have obvious limitations, deep learning is expected to become the basis of modern lncRNAX-miRNA interaction prediction algorithms.

关 键 词：lncRNA-miRNA相互作用 深度学习 数据库 计算框架 

分 类 号：TP18[自动化与计算机技术—控制理论与控制工程]