Huang Yufei, Tienda-Luna Isabel M, Wang Yufeng
Department of Electrical and Computer Engineering, University of Texas at San Antonio, San Antonio, TX 78249-0669,
IEEE Signal Process Mag. 2009 Jan 1;26(1):76-97. doi: 10.1109/MSP.2008.930647.
Statistical models for reverse engineering gene regulatory networks are surveyed in this article. To provide readers with a system-level view of the modeling issues in this research, a graphical modeling framework is proposed. This framework serves as the scaffolding on which the review of different models can be systematically assembled. Based on the framework, we review many existing models for many aspects of gene regulation; the pros and cons of each model are discussed. In addition, network inference algorithms are also surveyed under the graphical modeling framework by the categories of point solutions and probabilistic solutions and the connections and differences among the algorithms are provided. This survey has the potential to elucidate the development and future of reverse engineering GRNs and bring statistical signal processing closer to the core of this research.
本文对用于逆向工程基因调控网络的统计模型进行了综述。为了向读者提供该研究中建模问题的系统级视图，提出了一个图形化建模框架。该框架作为一个支架，在其上可以系统地整合对不同模型的综述。基于该框架，我们从基因调控的多个方面综述了许多现有模型；讨论了每个模型的优缺点。此外，还在图形化建模框架下，按照点解和概率解的类别对网络推理算法进行了综述，并给出了算法之间的联系和差异。这项综述有可能阐明逆向工程基因调控网络的发展和未来，并使统计信号处理更接近该研究的核心。
