A new strategy for assessing sensitivities in biochemical models

Area of Science:

• Systems Biology
• Computational Biology
• Pharmacology

Background:

• Systems biology relies on modeling and simulation, but parameter values are often unknown or unidentifiable.
• Sensitivity analysis quantifies parameter importance and can identify potential drug targets.
• Traditional sensitivity analysis methods are local and dependent on exact parameter values, which are often unknown.

Purpose of the Study:

• To develop an efficient global sensitivity analysis approach for systems biology models.
• To overcome the computational expense and limitations of local sensitivity analysis methods.
• To identify critical model parameters and potential drug targets by exploring a wide parameter space.

Main Methods:

• Utilized numerical optimization methods to search extensive parameter spaces.
• Determined the maximum and minimum values of metabolic control coefficients.
• Applied the strategy to a drug development example using COPASI software.

Main Results:

• Presented an efficient computational approach for global sensitivity analysis.
• Demonstrated the method's utility in identifying sensitive parameters and potential drug targets.
• Successfully applied the strategy in a relevant drug development context.

Conclusions:

• The proposed numerical optimization approach offers an efficient alternative for global sensitivity analysis in systems biology.
• This method effectively addresses the limitations of local sensitivity analysis, especially for high-dimensional systems.
• The strategy aids in identifying key parameters for model refinement and drug target discovery.