Alpha Synapse

[C*4]

Hello,

I am currently attempting to model an AMPA synapse as a point process. I used an existing model

(http://senselab.med.yale.edu/modeldb/Sh ... model=3808)
as an example and template for my code.

Code: Select all

TITLE Ampa2

INDEPENDENT {t FROM 0 TO 1 WITH 1 (ms)}

NEURON {
	POINT_PROCESS AmpaSyn
	RANGE onset, gMax, e, i, g, tau
	NONSPECIFIC_CURRENT i
}

UNITS {
	(nA) 	= (nanoamp)
	(mV)	= (millivolt)
	(nS) 	= (nanomho)
}

PARAMETER {
	onset		(ms)
	gMax = 0	(nS)
	tau = 18	(ms)
	e = 0.0		(mV)
	v		(mV)
}

ASSIGNED { 
	i 		(nA)  
	g 		(nS)
}

UNITSOFF


BREAKPOINT { LOCAL tt
	tt = (t-onset)
	if (t>onset) { 
	  g = gMax * exp(-tt/tau) * (1-exp(-tt/2.2))/0.68
	}
	else {g = 0}
	i = g * (v - e)
}
UNITSON

Now everything seems okay until I started playing with some values of gMax, and found that the resulting current values were a bit off.



Now if I'm correct with my calculations, I should have a current with an amplitude of -73pA, which doesn't seem to be the case. The g values seem to check out well, so I'm not sure what my problem is. Any suggestions?
Thanks!

[ted]

Now if I'm correct with my calculations, I should have a current with an amplitude of -73pA

What calculations did you do, and how did you do them?

The g values seem to check out well

How do you know that?

[Raj]

I think it would be wise to test the current in voltage clamp, so that you know what your actual electrochemical potential is. Especially when the synapse is inserted in a tiny compartment the induced depolarization will change the (v-e) term rapidly.

Note: The value of 0.68 is very specific for tau= 18 ms (I just checked in matlab), its occurence in its present form here is dangerous.

[C*4]

Now if I'm correct with my calculations, I should have a current with an amplitude of -73pA

What calculations did you do, and how did you do them?

The g values seem to check out well

How do you know that?

My calculation was plugging in the values into the equation
I = g*(V - Erev), with Erev = 0mV, V = -73mV and g = 1nS. That calculates out to a current of -73pA.



the graph of g appears to have a maximum value close to 1, which is the value of gMax in the point process.

[ted]

My calculation was plugging in the values into the equation
I = g*(V - Erev), with Erev = 0mV, V = -73mV and g = 1nS.

True, as long as v remains at -73 mV while g rises to 1 nS. This hypothesis is easily tested.

First, bring up a new "Current axis" graph (NEURON Main Menu / Graph / Current axis) and use the new graph's "Plot what?" to specify that it should plot AmpaSyn[0].g (the "Plot what? tutorial" is at http://www.neuron.yale.edu/neuron/docs/pwt/pwt.html). Be sure to return the new graph to Crosshair mode before proceeding further.

Next, change the synaptic mechanism's start time ("onset") to 1 ms, reduce Tstop to 10 ms, and run another simulation. You can discover the coordinates of plotted points by clicking on any graph trace and holding the button down while dragging the cursor back and forth along the length of the trace (as long as the graph is in Crosshair mode). At what time does g reach its maximum? What is the value of v at that time? How soon does v deviate by more than 10 mV from its initial value? At what time does synaptic current reach its maximum? What are the values of g and v at this time? Are i, g, and v consistent?