Hodgkin-Huxley revisited: reparametrization and identifiability analysis of the classic action potential model with...

Area of Science:

• Computational Biology
• Biophysics
• Physiology

Background:

• Complex cardiac cell models require robust parameterization methods.
• Loss of direct links between model parameters and experimental data hinders reproducibility and re-fitting.
• Approximate Bayesian computation (ABC) offers a potential solution for inferring parameter distributions.

Purpose of the Study:

• To evaluate the efficacy of ABC in parameter inference for the Hodgkin-Huxley cardiac action potential model.
• To assess the relationship between model parameters and experimental data.
• To establish a template for data-driven parametrization of new cardiac models.

Main Methods:

• Application of Approximate Bayesian Computation (ABC) to the Hodgkin-Huxley model.
• Inference of parameter distributions for ion channel gating rates and voltage dependencies.
• Comparison of ABC-derived parameters with original experimental values.

Main Results:

• ABC accurately reproduced reported values for sodium and potassium channel gating rates.
• Variable posteriors for voltage dependency parameters indicate insufficient constraint from voltage-clamp data alone.
• Hodgkin-Huxley parameter estimates remain competitive with ABC-derived values.

Conclusions:

• ABC is a viable method for parameter inference in complex cardiac models.
• Additional experimental data is necessary to fully constrain voltage dependency parameters.
• This study provides a foundation for cardiac model identifiability analysis and data-driven parametrization.