www.neuron.yale.edu

I am wondering how Neuron compute the impedance at different frequencies. I mean if I write the hoc code myself, Obviously I need to exert an input signal and then record the change of the membrane potential. For 0 HZ input impedance, it is easy to understand. What is the input signal for NEURON to compute the input impedance of a cell at 20 HZ for example. Is it a 20 Hz sinusoidal current wave? I am trying to simulate the efficacy of synaptic inputs at different parts of a dendrite to cause an EPSP at the soma. So transfer impedance should reflect this efficacy. However, I am not sure the transfer impedance of which frequency could best approximate this synaptic input efficacy. Thanks.

Impedance is calculated in the frequency domain from the Laplace transformation of the spatially discretized cable equation. As mentioned here
viewtopic.php?f=8&t=3515&p=14922
the second argument to the Impedance class's compute() method specifies whether voltage-dependent channel gating is taken into account, or whether all ion channels are treated as if they have time-invariant conductance.

The question of which frequency is most representative for a given synaptic input can only be answered empirically. Jaffe and Carnevale experimented with a range of frequencies until they found one that produced a spatial profile of transfer impedance that looked "most similar" to the plot of peak somatic epsp amplitude vs. synaptic distance from the soma.
Jaffe, D. & Carnevale, N.
Passive normalization of synaptic integration influenced by dendritic architecture
Journal of Neurophysiology, 1999, 82, 3268-3285.
You can expect that the "best" frequency will depend on the time course of synaptic conductance, but to get more specific you simply have to run a bunch of simulations and examine the results.

ted wrote:Impedance is calculated in the frequency domain from the Laplace transformation of the spatially discretized cable equation. As mentioned here
viewtopic.php?f=8&t=3515&p=14922
the second argument to the Impedance class's compute() method specifies whether voltage-dependent channel gating is taken into account, or whether all ion channels are treated as if they have time-invariant conductance.

The question of which frequency is most representative for a given synaptic input can only be answered empirically. Jaffe and Carnevale experimented with a range of frequencies until they found one that produced a spatial profile of transfer impedance that looked "most similar" to the plot of peak somatic epsp amplitude vs. synaptic distance from the soma.
Jaffe, D. & Carnevale, N.
Passive normalization of synaptic integration influenced by dendritic architecture
Journal of Neurophysiology, 1999, 82, 3268-3285.
You can expect that the "best" frequency will depend on the time course of synaptic conductance, but to get more specific you simply have to run a bunch of simulations and examine the results.

Thanks for the reply.