Ma Tianle, Zhang Aidong
Department of Computer Science and Engineering, University at Buffalo (SUNY), Buffalo, NY 14260-2500, United States.
Methods. 2017 Jul 15;124:36-45. doi: 10.1016/j.ymeth.2017.05.002. Epub 2017 May 18.
Reconstructing context-specific transcriptional regulatory network is crucial for deciphering principles of regulatory mechanisms underlying various conditions. Recently studies that reconstructed transcriptional networks have focused on individual organisms or cell types and relied on data repositories of context-free regulatory relationships. Here we present a comprehensive framework to systematically derive putative regulator-target pairs in any given context by integrating context-specific transcriptional profiling and public data repositories of gene regulatory networks. Moreover, our framework can identify core regulatory modules and signature genes underlying global regulatory circuitry, and detect network rewiring and core rewired modules in different contexts by considering gene modules and edge (gene interaction) modules collaboratively. We applied our methods to analyzing Autism RNA-seq experiment data and produced biologically meaningful results. In particular, all 11 hub genes in a predicted rewired autistic regulatory subnetwork have been linked to autism based on literature review. The predicted rewired autistic regulatory network may shed some new insight into disease mechanism.
重建特定背景下的转录调控网络对于解读各种条件下调控机制的原理至关重要。最近，重建转录网络的研究主要集中在个体生物体或细胞类型上，并依赖于无背景调控关系的数据存储库。在此，我们提出了一个综合框架，通过整合特定背景下的转录谱和基因调控网络的公共数据存储库，系统地推导任何给定背景下假定的调控因子-靶标对。此外，我们的框架能够识别全局调控电路中的核心调控模块和特征基因，并通过协同考虑基因模块和边（基因相互作用）模块，检测不同背景下的网络重连和核心重连模块。我们将我们的方法应用于分析自闭症RNA测序实验数据，并产生了具有生物学意义的结果。特别是，根据文献综述，预测的重连自闭症调控子网络中的所有11个枢纽基因都与自闭症有关。预测的重连自闭症调控网络可能为疾病机制提供一些新的见解。
