Modeling the excitability of mammalian nerve fibers: influence of afterpotentials on the recovery cycle

Encoded in CellML by Alan Garny

Auckland Bioengineering Institute, University of Auckland

Model Status

This model has been encoded using the information available in the published article. Only the nodal part of the model has been implemented:

• No equation to compute the membrane potential is provided. However, the model relies on the Hodgkin-Huxley formulation, so we rely on that formulation to compute the membrane potential. We use Cn as the capacitance and iNaf, iNap, iKs and iLk as the currents involved in the membrane dynamics. We also use iKf, but by default its conductance is set to 0 S/cm2. The authors mention that, based on experimental data, the value for gKf should be about 0.02 S/cm2, hence they tested their model using a range of values between 0.01 and 0.04 S/cm2.
• No initial values for the membrane potential (Vm) and the different gating variables (m, h, p, s and n) are provided. So, we set Vm to -80 mV (i.e. the resting potential used in the internodal segments) and the different gating variables to 0. From there, the model was run without any stimulus and for several days worth of simulation to reach steady state.
• No information about the kind of stimulus that should be applied to generate a default action potential is provided. However, various stimulus durations are mentioned in the published article, varying between 0.1 and 3 ms. So, we arbitrarily went for 0.5 ms. As for the stimulus amplitude, we went for the smallest value that yields an action potential (actually, a series of action potentials), i.e. 0.05 mA/cm2. (A stimulus amplitude of 0.04 mA/cm2 will just result in a blip.)

Output of running the SED-ML file in OpenCOR.