Bedaso Yosef, Bergmann Frank T, Choi Kiri, Medley Kyle, Sauro Herbert M
Department of Bioengineering, William H. Foege Building, Box 355061, Seattle, WA 98195-5061, USA.
BioQuant/COS, Heidelberg University, Heidelberg, Germany.
Biosystems. 2018 Jul;169-170:20-25. doi: 10.1016/j.biosystems.2018.05.008. Epub 2018 May 30.
The topology of a reaction network can have a significant influence on the network's dynamical properties. Such influences can include constraints on network flows and concentration changes or more insidiously result in the emergence of feedback loops. These effects are due entirely to mass constraints imposed by the network configuration and are important considerations before any dynamical analysis is made. Most established simulation software tools usually carry out some kind of structural analysis of a network before any attempt is made at dynamic simulation. In this paper, we describe a portable software library, libStructural, that can carry out a variety of popular structural analyses that includes conservation analysis, flux dependency analysis and enumerating elementary modes. The library employs robust algorithms that allow it to be used on large networks with more than a two thousand nodes. The library accepts either a raw or fully labeled stoichiometry matrix or models written in SBML format. The software is written in standard C/C++ and comes with extensive on-line documentation and a test suite. The software is available for Windows, Mac OS X, and can be compiled easily on any Linux operating system. A language binding for Python is also available through the pip package manager making it simple to install on any standard Python distribution. The bulk of the source code is licensed under the open source BSD license with other parts using as either the MIT license or more simply public domain. All source is available on GitHub (https://github.com/sys-bio/Libstructural).
反应网络的拓扑结构会对网络的动力学性质产生重大影响。这种影响可能包括对网络通量和浓度变化的限制,或者更隐蔽地导致反馈回路的出现。这些效应完全归因于网络结构所施加的质量限制,并且在进行任何动力学分析之前都是重要的考虑因素。大多数已有的模拟软件工具在进行动态模拟之前,通常会对网络进行某种结构分析。在本文中,我们描述了一个便携式软件库libStructural,它可以进行各种流行的结构分析,包括守恒分析、通量依赖性分析和枚举基本模式。该库采用了健壮的算法,使其能够用于具有两千多个节点的大型网络。该库接受原始的或完全标记的化学计量矩阵,或者以SBML格式编写的模型。该软件用标准C/C++编写,并配有大量的在线文档和一个测试套件。该软件适用于Windows、Mac OS X,并且可以在任何Linux操作系统上轻松编译。通过pip包管理器还可以获得Python语言绑定,这使得它可以在任何标准的Python发行版上轻松安装。大部分源代码遵循开源BSD许可,其他部分则使用MIT许可或更简单的公共领域许可。所有源代码均可在GitHub(https://github.com/sys-bio/Libstructural)上获取。