Kim Kyung Hyuk, Sauro Herbert M
Department of Bioengineering, University of Washington, William H. Foege Building, 355061, Seattle, WA, 98195, USA,
Methods Mol Biol. 2015;1244:287-97. doi: 10.1007/978-1-4939-1878-2_14.
This chapter introduces a computational analysis method for analyzing gene circuit dynamics in terms of modules while taking into account stochasticity, system nonlinearity, and retroactivity. (1) ANALOG ELECTRICAL CIRCUIT REPRESENTATION FOR GENE CIRCUITS: A connection between two gene circuit components is often mediated by a transcription factor (TF) and the connection signal is described by the TF concentration. The TF is sequestered to its specific binding site (promoter region) and regulates downstream transcription. This sequestration has been known to affect the dynamics of the TF by increasing its response time. The downstream effect-retroactivity-has been shown to be explicitly described in an electrical circuit representation, as an input capacitance increase. We provide a brief review on this topic. (2) MODULAR DESCRIPTION OF NOISE PROPAGATION: Gene circuit signals are noisy due to the random nature of biological reactions. The noisy fluctuations in TF concentrations affect downstream regulation. Thus, noise can propagate throughout the connected system components. This can cause different circuit components to behave in a statistically dependent manner, hampering a modular analysis. Here, we show that the modular analysis is still possible at the linear noise approximation level. (3) NOISE EFFECT ON MODULE INPUT-OUTPUT RESPONSE: We investigate how to deal with a module input-output response and its noise dependency. Noise-induced phenotypes are described as an interplay between system nonlinearity and signal noise. Lastly, we provide the comprehensive approach incorporating the above three analysis methods, which we call "stochastic modular analysis." This method can provide an analysis framework for gene circuit dynamics when the nontrivial effects of retroactivity, stochasticity, and nonlinearity need to be taken into account.
本章介绍了一种计算分析方法，该方法在考虑随机性、系统非线性和反馈作用的同时，从模块的角度分析基因回路动力学。（1）基因回路的模拟电路表示：两个基因回路组件之间的连接通常由转录因子（TF）介导，连接信号由TF浓度描述。TF被隔离到其特定的结合位点（启动子区域）并调节下游转录。已知这种隔离会通过增加TF的响应时间来影响其动力学。下游效应——反馈作用——已被证明在电路表示中可明确描述为输入电容增加。我们对此主题进行简要回顾。（2）噪声传播的模块化描述：由于生物反应的随机性，基因回路信号存在噪声。TF浓度的噪声波动会影响下游调控。因此，噪声可在相连的系统组件中传播。这可能导致不同的电路组件以统计相关的方式运行，妨碍模块化分析。在此，我们表明在线性噪声近似水平下模块化分析仍然可行。（3）噪声对模块输入 - 输出响应的影响：我们研究如何处理模块输入 - 输出响应及其噪声依赖性。噪声诱导的表型被描述为系统非线性和信号噪声之间的相互作用。最后，我们提供了结合上述三种分析方法的综合方法，我们称之为“随机模块化分析”。当需要考虑反馈作用、随机性和非线性的重要影响时，该方法可为基因回路动力学提供一个分析框架。
