Division of Molecular Biosciences, Imperial College London, London, UK.
Bioinformatics. 2012 Mar 1;28(5):731-3. doi: 10.1093/bioinformatics/btr714.
The growing interest in the role of stochasticity in biochemical systems drives the demand for tools to analyse stochastic dynamical models of chemical reactions. One powerful tool to elucidate performance of dynamical systems is sensitivity analysis. Traditionally, however, the concept of sensitivity has mainly been applied to deterministic systems, and the difficulty to generalize these concepts for stochastic systems results from necessity of extensive Monte Carlo simulations.
Here we present a Matlab package, StochSens, that implements sensitivity analysis for stochastic chemical systems using the concept of the Fisher Information Matrix (FIM). It uses the linear noise approximation to represent the FIM in terms of solutions of ordinary differential equations. This is the first computational tool that allows for quick computation of the Information Matrix for stochastic systems without the need for Monte Carlo simulations.
http://www.theosysbio.bio.ic.ac.uk/resources/stns
Supplementary data are available at Bioinformatics online.
人们对生化系统中随机性作用的兴趣日益浓厚，这促使人们需要开发工具来分析化学反应的随机动力学模型。灵敏度分析是阐明动力学系统性能的一种强大工具。然而，传统上，灵敏度的概念主要应用于确定性系统，并且将这些概念推广到随机系统的困难源于需要进行大量的蒙特卡罗模拟。
本文介绍了一个 Matlab 软件包 StochSens，它使用 Fisher 信息矩阵（FIM）的概念来对随机化学系统进行灵敏度分析。它使用线性噪声近似将 FIM 表示为常微分方程的解。这是第一个计算工具，它允许在不需要蒙特卡罗模拟的情况下快速计算随机系统的信息矩阵。
http://www.theosysbio.bio.ic.ac.uk/resources/stns
补充数据可在“Bioinformatics”在线获取。
