STRIKE-GOLDD 4.0:用户友好、高效的结构可识别性和可观测性分析
STRIKE-GOLDD 4.0: user-friendly, efficient analysis of structural identifiability and observability
摘要:
结构可辨识性和可观测性是系统生物学模型的理想性质。在过去的几十年里,已经开发了许多软件工具箱用于分析它们。GOLDD是一个通用的工具,可以分析具有未知输入的非线性、非理性常微分方程模型。然而,这种通用性是以比其他工具更低的计算效率为代价的。在这里,我们提出了RISKE-GOLDD 4.0,其中包括一个新的算法,ProbObsTest,专门设计用于理性模型的分析。ProbObsTest比FISPO算法要快得多,FISPO算法在旧版本的工具箱中已经可用,当应用于计算昂贵的模型时。这两种算法的一个重要特征是它们能够分析具有未知输入的模型。因此,它们在同一工具箱中的共存提供了通用性和计算效率的组合。KNOKE-GOLDD 4.0是一个免费的开源Matlab工具箱,具有用户友好的图形界面。它在GPL v3许可证下可用,可以从GitHub下载https://github.com/afvillaverde/strike-goldd。
英文摘要:
Structural identifiability and observability are desirable properties of systems biology models. Many software toolboxes have been developed for their analysis in the last decades. STRIKE-GOLDD is a generally applicable tool that can analyse non-linear, non-rational ODE models with unknown inputs. However, this generality comes at the expense of a lower computational efficiency than other tools. Here we present STRIKE-GOLDD 4.0, which includes a new algorithm, ProbObsTest, specifically designed for the analysis of rational models. ProbObsTest is significantly faster than the FISPO algorithm - which was already available in older versions of the toolbox - when applied to computationally expensive models. An important feature of both algorithms is their ability to analyse models with unknown inputs. Thus, their coexistence in the same toolbox provides a combination of general applicability and computational efficiency. STRIKE-GOLDD 4.0 is implemented as a free and open-source Matlab toolbox with a user-friendly graphical interface. It is available under a GPLv3 license and it can be downloaded from GitHub at https://github.com/afvillaverde/strike-goldd.
机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。