Reducing experimental variability in variance-based sensitivity analysis of biochemical reaction systems

Area of Science:

• Systems Biology
• Computational Biology
• Biochemistry

Background:

• Sensitivity analysis is crucial for understanding biological variability in cellular behavior.
• Existing methods struggle with experimental uncertainty, limiting their practical application in systems biology.
• Accurate nominal parameter values are often unattainable due to experimental variability.

Purpose of the Study:

• To develop a probabilistic approach for sensitivity analysis of biochemical reaction systems.
• To explicitly model and reduce the impact of experimental variability on sensitivity analysis results.
• To provide a robust technique for systems biology problems with significant experimental uncertainty.

Main Methods:

• Proposed a probabilistic approach integrating a variance-based method for sensitivity analysis.
• Employed numerical techniques including Monte Carlo estimation, derivative approximation, and dimensionality reduction.
• Utilized a computational model of the epidermal growth factor receptor signaling pathway for validation.

Main Results:

• The proposed technique significantly reduces the influence of experimental variability on sensitivity analysis outcomes.
• Demonstrated effectiveness using a model of the epidermal growth factor receptor signaling pathway.
• The method accommodates appreciable levels of experimental variability.

Conclusions:

• The novel probabilistic approach offers substantial improvements over existing sensitivity analysis techniques.
• This method enhances the reliability and applicability of sensitivity analysis in the presence of experimental uncertainty.
• The approach is particularly valuable for complex biochemical reaction systems in systems biology.