几类典型基因调控网络的混合建模及动力学分析

Systems biology is included in the major strategic needs of the country's basic research fields in the national medium and long-term science and technology planning, and it is also treated as the prior to the deployment of development areas in department of information science in the CSC twelfth five-year development plan. Gene regulatory networks are very important complex biological networks in the field of systems biology. However, current investigations mainly focus on discrete Boolean modeling and differential equation modeling of simple circuits,few models and quantitative analysis results have been reported on network-level gene regulatory networks, which are mainly due to its structural complexicity and the complexicity of differential equation models of network-level gene regulatory networks. Based on complex networks and dynamical system theory, this project aims to investigate the hybrid discrete boolean and differential equation modeling and dynamical analysis of several representative gene regulatory networks, which include simple circuits、intergrated circuits、coupled multi-cellular circuits as well as some general medium-sized and large-sized gene regulatory networks;Through the modeling and dynamical analysis of serveral simple-to-complex gene networks, this project expects to promote system-level investigations of systems biology, and to provide method to quantatively clarify the stimulus-response relationships and the design principles of complex gene regulatory networks. Related investigations will help to systemly understand the cellular activity and life at the genome level, which are promising in explaining the occurrence、the development and the treatment of disease, and they also have important theoretical value and pontential applications in networked medincine、personalized medincine and synthetic biology and so on.

系统生物学是国家中长期科技规划中面向国家重大战略需求的基础研究领域，又是基金委十二五发展规划纲要中信息科学部优先部署发展的领域。基因调控网络是系统生物学研究的一类很重要的复杂生物网络。但是，目前主要集中于简单环路的离散布尔模型及微分方程建模研究，由于网络层次基因调控网络自身的结构复杂性及微分方程模型的复杂性，复杂基因调控网络的定量建模及分析还很少。基于复杂网络及动力系统理论，本项目旨在研究几类典型基因调控网络的混合布尔和微分方程建模及动力学分析，包括简单环路、整合环路、多细胞耦合环路及一般的大中型调控网络；通过对若干从简单到复杂的基因网络的建模分析，推动系统层次的系统生物学研究，为量化的阐明复杂的基因调控网络的刺激-响应关系及设计原理提供思路。相关研究有助于在基因组层次上系统的理解细胞及生命活动，解释疾病的发生、发展以及治疗，对网络医学、个性化医学和合成生物学等具有重要理论价值和应用前景。

本项目基于复杂网络及动力系统理论，围绕基因调控网络的定量分析，取得了如下成果：首先，我们研究了若干简单基因调控环路的结构与动力学、功能之间的关系，揭示了这些环路的设计原理。如我们研究了有色噪声介导的双稳切换以及和白噪声诱导行为的区别；研究了多重正反馈环路中附加正反馈的作用，解释了为何实际系统中频繁出现多重正反馈机制。第二，我们设计了若干新的表征网络中节点重要性的指标。如通过修正LeaderRank, 提出一种更具鲁棒性的自适应算法；我们还设计了一种基于邻居信息收集的算法，包括已有的度、半局部中心度、LeaderRank等八种算法作为其特殊情形，其极限情形对应于特征向量中心度；此外，基于网络模体及主成分分析，提出了一种在有向及无向生物分子网络中识别关键节点的算法。第三，基于复制-变异的机制，通过计算机仿真及统计分析，阐明了生物分子网络中一些网络模体的演化机制，有助于人们理解复杂生物系统的进化过程。最后，我们探讨了理论研究的若干生物医学应用。如通过与生命科学研究者合作，从实验测序获得的油菜转录组数据中识别出169个关键抗旱基因，并对这些基因进行了详细的生物学研究。通过对若干从简单到复杂的基因网络的定量分析，推动了系统层次的系统生物学研究，为量化的阐明复杂的基因调控网络的刺激-响应关系及设计原理提供了思路。相关研究有助于在组学层次上系统的理解细胞及生命活动，对网络医学、合成生物学和作物新品种培育等具有重要理论价值和应用前景。