A very simple Voltage-clamp experiment

The basics of how to develop, test, and use models.
Post Reply
Nin
Posts: 41
Joined: Thu May 03, 2007 4:04 pm
Location: Institute of Science and Technology (IST Austria)
Contact:

A very simple Voltage-clamp experiment

Post by Nin »

Dear forum members,

I am trying to study the effect of the series resistance on time course of synaptic currents. For that, I first created a very simplified model of a soma and an "alpha synapse" attached to it . To perform the measurements, I would like to voltage-clamp the soma at -70mV and get the corresponding postsynaptic currents.

After creating a simplified model (see code bellow), I could see the corresponding synaptic potential, but unfortunately, I could not access the membrane current (Graph->Current Axis->Plot What->soma.i_pass(.5) does not show anything). How do I see the corresponding synaptic current?

I tried to perform the V-Clamp experiment (SEVClamp with rs=0 and vc = -70) but this did not changed anything.
Is there any way to access the i or g variables that are shown in the menu to get the synaptic current?

Any help to would be greatly appreciated. Thanks a lot in advance.

Code: Select all

/* simpletest.hoc */
load_file("nrngui.hoc")

//-------------------------------------------------------------------------
// initial simulation parameters
//-------------------------------------------------------------------------
t = 0
dt = 0.005 
tstop = 150 
Vrest =-70
v_init = Vrest

//-------------------------------------------------------------------------
// Basic morphology/biophysics
//-------------------------------------------------------------------------

Ri = 294
Rm = 164000
Cm = 0.683

create soma
soma  {nseg = 3 L = 30 diam = 30 }

forall {insert pas g_pas = 1/Rm e_pas = Vrest}
forall {v= Vrest Ra = Ri cm = Cm}


//-------------------------------------------------------------------------
// Alpha synapse
//-------------------------------------------------------------------------

access soma
objectvar mySyn 
mySyn = new AlphaSynapse(0.5)

mySyn.onset = 20
mySyn.tau = 0.3
mySyn.gmax = 0.008
mySyn.e    = -5
ted
Site Admin
Posts: 5810
Joined: Wed May 18, 2005 4:50 pm
Location: Yale University School of Medicine
Contact:

Re: A very simple Voltage-clamp experiment

Post by ted »

soma.i_pass(.5) does not show anything
There is no such variable as i_pass because there is no mechanism with SUFFIX pass. Perhaps you meant i_pas?
How do I see the corresponding synaptic current?
The AlphaSynapse mechanism has a variable i that reports the current that it produces (see Programmer's Reference documentation on AlphaSynapse).

If you want to see membrane current, you should be looking at total membrane current (ionic + capacitive current), not just the current that passes through the pas mechanism (i.e. current that flows through ion channels). Insert the extracellular mechanism into your model, and the total membrane current of any segment will be available as i_membrane(x) where 0<x<1 is the range of a point that is contained in the segment (see documentation of extracellular in the Programmer's Reference).

Time for some hints.

1. Make your life easier by simplifying the model specification. The model cell has diameter = length. With this form factor, and Ra and cm values that are even remotely "reasonable", there is no need for nseg to be > 1.
2. Note that i_membrane will be in units of mA/cm2, i.e. a current density, so to get net membrane current in nA it is necessary to multiply by area(0.5)*1e-8. area(0.5) is the area of the entire section if nseg = 1, and has units of square microns. The 1e-8 is a scale factor that is necessary to reconcile the difference in magnitude between square microns and square cm.
3. Make your life even easier by using reasonable run times. The AlphaSynapse generates a current waveform that reaches a peak at t = tau, and decays at a rate that, as t becomes large, can be approximated by an exponential with time constant tau. At times larger than 5*tau the current will be extremely small.

The other time constant in the system is what I will call tau_x. The value of tau_x equals the product
net membrane capacitance * Rx
where Rx = parallel combination of net membrane resistance and SEClamp series resistance. For any series resistance up to 1 gigohm, tau_x will be so fast that any perturbation of membrane potential will vanish in less than 10 ms.

Finally, if the model cell has a resting potential of -70 mV, and the SEClamp's holding potential is also -70 mV, there is no need for a long delay before activating the synapse.

So set the AlphaSynapse's onset to 1 ms, make tstop 10 ms, let dt = 0.025 ms, and you'll have faster runs, generate just enough data for it to be useful, and produce fewer meaningless numbers that have to be ignored.
I tried to perform the V-Clamp experiment (SEVClamp with rs=0 and vc = -70)
It's called SEClamp. The clamp current is one of the variables that is reported by the SEClamp mechanism (see Programmer's Reference etc.).
Nin
Posts: 41
Joined: Thu May 03, 2007 4:04 pm
Location: Institute of Science and Technology (IST Austria)
Contact:

Re: A very simple Voltage-clamp experiment

Post by Nin »

Dear Ted,

thanks a lot for your fabulous and advices. The basic model started to work nicely and I started the changes to perform more complicated simulation I am particularly thankful for your *hints* since this is something that we cannot get in the books and manuals, and required a more experienced person.

Thanks!
Post Reply