Identifying enhancer-promoter interactions with neural network based on pre-trained DNA vectors and attention mechanism
Identifying enhancer-promoter interactions with neural network based on pre-trained DNA vectors and attention mechanism
复制标题
基于预先训练的 DNA 向量和注意力机制识别增强子-启动子与神经网络的相互作用
DOI：
10.1093/bioinformatics/btz694
复制
发表时间：
2020-02-15
期刊：
影响因子：
5.8
通讯作者：
Liu, Xiangrong
中科院分区：
文献类型：
--
作者：
Hong, Zengyan;Zeng, Xiangxiang;Liu, Xiangrong
动机：确定增强子-启动子相互作用(EPI)对人类发育具有重要意义。然而，鉴定EPI的实验方法耗费了太多的时间、人力和金钱。因此，越来越多的研究致力于开发计算方法来解决这一问题。不幸的是，大多数现有的计算方法需要各种基因组数据，而这些数据并不总是可用的，特别是对于新的细胞系。因此，它限制了方法的大规模实际应用。结果：在本文中，我们提出了一种新的深度学习方法，即EPIVAN方法，该方法能够仅使用基因组序列来预测远程EPIS。为了探索关键序列特征，我们首先使用预先训练好的DNA载体编码增强子和启动子；然后使用一维卷积和门控递归单元来提取局部和全局特征；最后利用注意力机制来提高关键特征的贡献度，进一步提高EPIVAN的性能。在六个细胞系上的基准比较表明，EPIVAN比最先进的预测者表现得更好。此外，我们还建立了一个通用的模型，该模型具有转移能力，可以用来预测不同细胞系的EPI。
Motivation: Identification of enhancer-promoter interactions (EPIs) is of great significance to human development. However, experimental methods to identify EPIs cost too much in terms of time, manpower and money. Therefore, more and more research efforts are focused on developing computational methods to solve this problem. Unfortunately, most existing computational methods require a variety of genomic data, which are not always available, especially for a new cell line. Therefore, it limits the large-scale practical application of methods. As an alternative, computational methods using sequences only have great genome-scale application prospects.Results: In this article, we propose a new deep learning method, namely EPIVAN, that enables predicting long-range EPIs using only genomic sequences. To explore the key sequential characteristics, we first use pre-trained DNA vectors to encode enhancers and promoters; afterwards, we use one-dimensional convolution and gated recurrent unit to extract local and global features; lastly, attention mechanism is used to boost the contribution of key features, further improving the performance of EPIVAN. Benchmarking comparisons on six cell lines show that EPIVAN performs better than state-of-the-art predictors. Moreover, we build a general model, which has transfer ability and can be used to predict EPIs in various cell lines.