Related Experiment Videos
Quantifying robustness of biochemical network models
1Department of Electrical and Computer Engineering, The Johns Hopkins University, Baltimore, MD USA. lma@jhu.edu
BMC Bioinformatics
|December 17, 2002
Summary
Quantifying the robustness of biochemical models is crucial for validation. This study introduces structural singular value (SSV) analysis, revealing poor robustness in oscillatory models when parameters vary.
Area of Science:
- Systems Biology
- Biochemical Network Modeling
Background:
- Model validation and selection rely on understanding the robustness of mathematical models for biochemical networks.
- Quantifying parametric robustness is essential but currently lacks adequate tools.
Purpose of the Study:
- To present and contrast two quantitative techniques for assessing the robustness of oscillatory biochemical models.
- To evaluate the utility of structural singular value (SSV) analysis in this context.
Main Methods:
- Employed single-parameter bifurcation analysis to assess limit cycle oscillation stability, frequency, and amplitude.
- Utilized structural singular value (SSV) analysis from control engineering to quantify robust stability of the limit cycle.
Main Results:
- Single-parameter bifurcation analysis provided insights into stability and oscillation characteristics.
- SSV analysis indicated very poor robustness when model parameters were allowed to vary.
- A comparison highlighted the strengths of each method in assessing model robustness.
Conclusions:
- Incorporating SSV analysis alongside single-parameter sensitivity analysis offers a powerful approach to quantify model robustness.
- The findings underscore the importance of robust parameter estimation and model validation in systems biology.