Current Meaasurment at the soma

[Franzi]

Hello,

i want to know if there is any possibility to measure the whole current at the soma. I think with

vector.record(&soma.ina(0.5)), i can measure the NA+-current

vector.record(&soma.ica(0.5)), i can measure the CA2+-current and with

vector.record(&soma.ik(0.5)) i can measure the K+-current.

But I want to measure the whole current.

2nd, i manipulated the e_pas and g_pas value of dendrites of a neuron, then measuring the voltage at the middle of the soma. Until a specific value of e_pas and g_pas the voltage is as good as i expected (it produces nice regular spike trains), but then, if the g_pas (electric conductivity) and e_pas is too good, neuron calculates voltage values, they are not within the biophysical constrains (e.g. -8000 to 1e+25 and so on). The first spike is good, but then the voltage values break out. Is this appearance known? How can i avoid this? Note that we calculate the e_pas and g_pas value in such a way that the background activity firing rate of some alpha synapse is correct simulated, that means, if any background activity is present at some dendrite, the g_pas value of this dendrite rises.

Thank you for help.

Greetings, Franzi

[ted]

i want to know if there is any possibility to measure the whole current at the soma.

Use extracellular, which gives you i_membrane
http://www.neuron.yale.edu/neuron/stati ... racellular
soma insert extracellular
Don't forget that i_membrane is a range variable, and its units are density units.
So if soma has nseg>1, you must sum the product of i_membrane and area over all
internal nodes. For example,

Code: Select all

run()
soma for (x,0) print i_membrane(x)*area(x)

will give you total soma membrane current at the end of a simulation. So you'll want to
record from each internal node

Code: Select all

objref veclist, tempvec
soma for (x,0) {
  tempvec = new Vector()
  tempvec.record(&i_membrane(x))
  veclist.append(tempvec)
}
objref tempvec

and at the end of a simulation

Code: Select all

i = 0
soma for (x,0) {
  veclist.o(i).mul(area(x))
  i += 1
}

Then you'll want to add the vectors

Code: Select all

objref itotal
itotal = new Vector(veclist.o(0).size(), 0)
for i = 0, veclist.count()-1 itotal.add(veclist.o(i))

As far as what happens when you adjust model parameters, I can't even begin to guess
unless I can reproduce the phenomenon.

[Franzi]

Hello ted,

your hints were very useful, it will help me.

The second problem i wrote is solved: In one formula for g_pas a part of that g_pas was negative (i swapped A-B into B-A by mistake). A negative g_pas goes without saying undefined.

Greetings,
Franzi.