Easiest way to measure conductance

The basics of how to develop, test, and use models.
Post Reply
etano

Easiest way to measure conductance

Post by etano »

I want to measure the peak conductance at the soma of a single neuron. I was thinking I should voltage clamp the soma, present the stimulus, and then measure the peak current. From there I could calculate the peak conductance. Should I use VClamp or SEClamp? Or is there some way just to read it out?
ted
Site Admin
Posts: 5810
Joined: Wed May 18, 2005 4:50 pm
Location: Yale University School of Medicine
Contact:

Re: Easiest way to measure conductance

Post by ted »

"the peak conductance" of what? A channel, a synapse, total conductance of the soma, or 1/input resistance of the cell? If one of the last two, does this model have > 1 compartment?
etano

Re: Easiest way to measure conductance

Post by etano »

Sorry for not being more specific. I would like the total conductance of the soma, and the model has >1 compartment.
ted
Site Admin
Posts: 5810
Joined: Wed May 18, 2005 4:50 pm
Location: Yale University School of Medicine
Contact:

Re: Easiest way to measure conductance

Post by ted »

etano wrote:I would like the total conductance of the soma, and the model has >1 compartment.
That's helpful information.
I was thinking I should voltage clamp the soma, present the stimulus, and then measure the peak current. From there I could calculate the peak conductance.
Nope. Imagine a voltage clamped cell with multiple voltage-gated channels. To make the problem simple, assume that the cell is a sphere. Apply a depolarizing step command. Some currents are outward, others inward, and each has its own time course. Total ionic current is area * SUMMA gi(t)*(v(t) - ei) where each current corresponds to a different i. Total conductance is SUMMA gi(t), which does not follow the same time course as total ionic current. If the cell is not a sphere, the problem is even more complex because v and gi(t) are not uniform over the surface of the cell (not to mention the inability of a voltage clamp at the soma to capture all membrane current from membrane that is not in the same compartment).

Here's what you need to do:
1. Make sure that each channel in the soma, whether it belongs to a density mechanism or a point process, has a conductance variable that is visible at the hoc level. This may require revision of whatever mod files you are using.
2. Use the Vector class's record() method to capture the time course of each conductance in each segment of the soma.
3. At the end of the simulation define a new Vector that will hold the total somatic conductance time course.
For each segment of the soma
--calculate the time course of the sum of all density mechanism ionic conductances
--multiply this vector by segment area
--add this vector to the total somatic conductance time course vector
Then add up the time courses of all of the somatic point process ionic conductances, and add that vector to the total somatic conductance time course vector.
etano

Re: Easiest way to measure conductance

Post by etano »

Alright, this works, well, for the the most part. It works for one model of mine where I am using templates and record with the following syntax:

Code: Select all

objref caVector
caVector = new Vector()
caVector.record(&cell[nCell].soma.gca_ca)
However in a much simpler model, I keep getting the same error:

Code: Select all

/usr/local/nrn/i686/bin/nrniv: syntax error
in demofig1.hoc near line 382
  caVector.record(&soma.gca_ca)
with the following syntax:

Code: Select all

objref caVector
caVector = new Vector()
caVector.record(&soma.gca_ca)
Clearly I am doing something wrong, but I'm not sure where my error is. Please help. Thanks ahead of time.
ted
Site Admin
Posts: 5810
Joined: Wed May 18, 2005 4:50 pm
Location: Yale University School of Medicine
Contact:

Re: Easiest way to measure conductance

Post by ted »

Vector record usually requires specification of the location at which to record the range variable, e.g.
caVector.record(&soma.gca_ca(0.5))
I am surprised that this
caVector.record(&cell[nCell].soma.gca_ca)
didn't produce an error message.
Post Reply