Deep learning of the tissue-regulated splicing code.
Deep learning of the tissue-regulated splicing code.
复制标题
DOI：
10.1093/bioinformatics/btu277
复制
发表时间：
2014-06-15
期刊：
影响因子：
--
通讯作者：
Frey BJ
中科院分区：
文献类型：
--
作者：
Leung MK;Xiong HY;Lee LJ;Frey BJ
动机：选择性剪接（AS）是一个受调控的过程，指导从单个基因产生不同的转录本。一个计算模型，可以准确地预测剪接模式的基因组特征和细胞背景的基础上是非常可取的，无论是在理解这种普遍的现象，并在探索遗传变异对AS的影响。研究方法：使用深度神经网络，我们开发了一个从小鼠RNA-Seq数据推断的模型，该模型可以预测单个组织中的剪接模式以及组织间剪接模式的差异。我们的架构使用隐藏变量，这些变量在进行预测时共同代表基因组序列和组织类型中的特征。使用图形处理单元来大大减少我们具有数百万参数的模型的训练时间。结果：我们表明，深层架构超过了以前的贝叶斯方法预测AS模式的性能。通过适当的优化过程和超参数的选择，我们证明了深度架构可以是有益的，即使是适度稀疏的数据集。提出了模型在基因组特征方面所学到的东西的分析。联系方式：frey@psi.toronto.edu补充信息：补充数据可从生物信息学在线网站获得。
Motivation: Alternative splicing (AS) is a regulated process that directs the generation of different transcripts from single genes. A computational model that can accurately predict splicing patterns based on genomic features and cellular context is highly desirable, both in understanding this widespread phenomenon, and in exploring the effects of genetic variations on AS. Methods: Using a deep neural network, we developed a model inferred from mouse RNA-Seq data that can predict splicing patterns in individual tissues and differences in splicing patterns across tissues. Our architecture uses hidden variables that jointly represent features in genomic sequences and tissue types when making predictions. A graphics processing unit was used to greatly reduce the training time of our models with millions of parameters. Results: We show that the deep architecture surpasses the performance of the previous Bayesian method for predicting AS patterns. With the proper optimization procedure and selection of hyperparameters, we demonstrate that deep architectures can be beneficial, even with a moderately sparse dataset. An analysis of what the model has learned in terms of the genomic features is presented. Contact: frey@psi.toronto.edu Supplementary information: Supplementary data are available at Bioinformatics online.
登录
查看更多内容
影响因子：
48
作者：
Katz, Yarden;Wang, Eric T.;Airoldi, Edoardo M.;Burge, Christopher B.
通讯作者：
Burge, Christopher B.
DOI：
10.1111/j.1467-9868.2011.00771.x
发表时间：
2011-01-01
影响因子：
5.8
作者：
Tibshirani, Robert
通讯作者：
Tibshirani, Robert
影响因子：
5.8
作者：
Di Lena, Pietro;Nagata, Ken;Baldi, Pierre
通讯作者：
Baldi, Pierre
影响因子：
32.8
作者：
Bengio, Yoshua
通讯作者：
Bengio, Yoshua
影响因子：
5.8
作者：
Xiong, Hui Yuan;Barash, Yoseph;Frey, Brendan J.
通讯作者：
Frey, Brendan J.