细胞信号转导网络输入信号的最优实验设计  

Optimal experimental design for input signal of cell signal transduction networks

作　　者：贾建芳[1] 刘太元[1] 岳红[1] 王宏[1] 

机构地区：[1]中国科学院自动化研究所

出　　处：《计算机与应用化学》2007年第8期1033-1038,共6页Computers and Applied Chemistry

基　　金：中国科学院海外杰出学者基金(2004-1-4)

摘　　要：细胞信号转导网络的结构复杂,规模庞大,具有高度非线性,数学模型的参数和变量数目很多,有效估计未知参数是系统生物学主要的研究问题之一。提出以Fisher信息矩阵和估计参数的协方差矩阵为基础,利用不同的优化实验设计方法,研究NF-κB信号转导网络输人信号IKK初浓度的最优设计问题,分析模型参数的置信区间,为参数估计设计优化实验。仿真结果表明,采用E-最优实验设计准则,获得输入信号IKK初浓度为0.06μM,能够提供较多的实验信息,提高参数估计的精度。Cell signal transduction networks have complex structure and large size. A mathematical model of cell signal transduction networks is highly nonlinear, and it includes a lot of parameters and variables. How to effectively estimate the unknown parameters is a major problem of systems biology. Based on Fisher information matrix and eovarianee matrix of estimated parameters, optimal experimental design for the initial concentration of input signal IKK of NF-KB signaling pathway was studied by means of various criterions. The confidence interval of parameters was analyzed and optimal experiment for parameter estimation was proposed. According to the Eoptimal experimental design criterion, the initial concentration of IKK is set to 0. 06μM. The simulation results demonstrate that more information for parameter estimation is obtained and the accuracy of parameter estimation is improved evidently.

关 键 词：最优实验设计 Fisher信息矩阵 细胞信号转导网络 参数估计 

分 类 号：Q332[生物学—遗传学]