An efficient and very accurate method for calculating steady-state sensitivities in metabolic reaction systems

Fumihide Shiraishi, Erika Yoshida, Eberhard O. Voit

Research output: Contribution to journalArticle

9 Citations (Scopus)

Abstract

Stability and sensitivity analyses of biological systems require the ad hoc writing of computer code, which is highly dependent on the particular model and burdensome for large systems. We propose a very accurate strategy to overcome this challenge. Its core concept is the conversion of the model into the format of biochemical systems theory (BST), which greatly facilitates the computation of sensitivities. First, the steady state of interest is determined by integrating the model equations toward the steady state and then using a Newton-Raphson method to fine-tune the result. The second step of conversion into the BST format requires several instances of numerical differentiation. The accuracy of this task is ensured by the use of a complex-variable Taylor scheme for all differentiation steps. The proposed strategy is implemented in a new software program, COSMOS, which automates the stability and sensitivity analysis of essentially arbitrary ODE models in a quick, yet highly accurate manner. The methods underlying the process are theoretically analyzed and illustrated with four representative examples: a simple metabolic reaction model; a model of aspartate-derived amino acid biosynthesis; a TCA-cycle model; and a modified TCA-cycle model. COSMOS has been deposited to https://github.com/BioprocessdesignLab/COSMOS.

Original languageEnglish
Article number6857342
Pages (from-to)1077-1086
Number of pages10
JournalIEEE/ACM Transactions on Computational Biology and Bioinformatics
Volume11
Issue number6
DOIs
Publication statusPublished - Nov 1 2014

All Science Journal Classification (ASJC) codes

  • Biotechnology
  • Genetics
  • Applied Mathematics

Fingerprint Dive into the research topics of 'An efficient and very accurate method for calculating steady-state sensitivities in metabolic reaction systems'. Together they form a unique fingerprint.

  • Cite this