Global sensitivity analysis for developing biological models: Application to K+ channel model in mouse ventricular...

Area of Science:

• Computational biology
• Biophysics
• Mathematical modeling

Background:

• Cardiac myocyte models are complex and nonlinear, with numerous parameters.
• Parameter estimation from noisy experimental data introduces uncertainty, impacting model reliability.
• Parametric uncertainty can significantly affect the accuracy and credibility of model predictions.

Purpose of the Study:

• To develop and apply a global sensitivity analysis method to quantify parametric uncertainty in cardiac myocyte models.
• To identify and rank the most influential parameters affecting model responses.
• To improve the accuracy and reliability of nonlinear K+ channel models.

Main Methods:

• Implementation of a global sensitivity analysis technique using Sobol indices.
• Application of the method to nonlinear K+ channel models of mouse ventricular myocytes.
• Evaluation of the impact of parametric uncertainty on model outputs.

Main Results:

• The developed global sensitivity analysis effectively determines the significance of parametric uncertainty.
• Key parameters influencing the responses of mouse ventricular myocyte K+ channel models were identified and ranked.
• The efficacy of the algorithm in handling uncertainty propagation was demonstrated.

Conclusions:

• Global sensitivity analysis with Sobol indices is a powerful tool for understanding parametric uncertainty in complex biological models.
• Reducing uncertainty in sensitive parameters is essential for enhancing the predictive power of cardiac myocyte models.
• This approach provides a framework for improving the credibility of computational models in cardiovascular research.