Extracellular Stimulation

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

Extracellular Stimulation

Post by nevo »

I have just begun to learn to use NEURON and I already read the comments to extracellular stimulation but I still have some questions.
I want to simulate voltage-controlled extracellular stimulation to the Spinal Motor Neuron Model (McIntyre et al 2002), which uses fast Na+, persistent Na+ and slow K+ channels. In this model the extracellular mechanism is inserted to each compartment. Within the node compartments the values amounts xraxial= 5.8574e5 xg=1e10 xc=0 for a fiber Diameter of 5.7 µm.
I calculated the extracellular potential values using a Finite Element Model. There exist values from -900 to 0 mV. Is it possible to use the “ecstim” program to apply the potential values to each compartment? What kind of unit and value is the threshold value in the program to elicit a spike?
What other alternatives do I have to apply extracellular stimulation to the Axon Model? Any help would be appreciated.
Thanks
ted
Site Admin
Posts: 5810
Joined: Wed May 18, 2005 4:50 pm
Location: Yale University School of Medicine
Contact:

Re: Extracellular Stimulation

Post by ted »

nevo wrote:I have just begun to learn to use NEURON and I already read the comments to extracellular stimulation but I still have some questions.
I want to simulate voltage-controlled extracellular stimulation to the Spinal Motor Neuron Model (McIntyre et al 2002)
You refer to McIntyre & Grill "Extracellular stimulation of central neurons . . . " JNP 88:1592-1604, 2002? Or their axon model, described in "Modeling the excitability of mammalian nerve fibers: influence of afterpotentials on the recovery cycle." JNP 87:995-1006, 2002?
I calculated the extracellular potential values using a Finite Element Model. There exist values from -900 to 0 mV. Is it possible to use the “ecstim” program to apply the potential values to each compartment?
So, instead of using an "activation function" approach (injecting current into each compartment to emulate the effect of extracelluar stimulation), which is what McIntyre & Grill did with their motoneuron model, you want to actually change the extracellular potential adjacent to each compartment as illustrated by the code in http://www.neuron.yale.edu/ftp/ted/neur ... nd_rec.zip? The answer is yes.
What kind of unit and value is the threshold value in the program to elicit a spike?
It will be a current, whose units and magnitude will depend on the anatomical and biophysical properties of your model cell, the conductivity of the medium in which it and the stimulating electrode(s) are embedded, and the positions and orientations of the model cell and the electrode(s).
nevo

Re: Extracellular Stimulation

Post by nevo »

I referred to the axon model described in "Modeling the excitability of mammalian nerve fibers: influence of afterpotentials on the recovery cycle." JNP 87:995-1006, 2002.
ted wrote:…you want to actually change the extracellular potential adjacent to each compartment as illustrated by the code in http://www.neuron.yale.edu/ftp/ted/neur ... nd_rec.zip.
Yes, but as I already get the potential values from my finite element model, I tried to use ecstim.zip. The comment of stim.hoc denotes “//drive e_extracellular at each internal node of the model with a voltage that has the time course specified by pvec, tvec”. So I assumed the value of pvec is a voltage value and wondered if it is mV or V.

When I use the ecstim program with the fast Na+, persistent Na+ and slow K+ channels and set the value for pvec to zero I nevertheless get an action potential (duration 3.5 ms). Can you explain me the reason for that?

Thanks a lot for your help.
ted
Site Admin
Posts: 5810
Joined: Wed May 18, 2005 4:50 pm
Location: Yale University School of Medicine
Contact:

Re: Extracellular Stimulation

Post by ted »

ecstim.zip contains only a toy demonstration of the principle of how to use the extracellular mechanism to implement extracellular stimulation. Being a toy demonstration, it assumes the very simplest case: that extracellular potential is a linear function of distance, and that the neural structure is represented by a single section. Substantial code revision is required if either of these assumptions is violated, as must be the case with the McIntyre & Grill axon (which has multiple sections). And is the extracellular potential generated by your finite element model a linear function of distance? Probably not.

If you still want to use the ecstim approach, you must generate a separate vector of calculated extracellular potentials for each compartment of your model, and drive each compartment's e_extracellular with its own vector.
I assumed the value of pvec is a voltage value and wondered if it is mV or V.
mV
When I use the ecstim program with the fast Na+, persistent Na+ and slow K+ channels and set the value for pvec to zero I nevertheless get an action potential (duration 3.5 ms). Can you explain me the reason for that?
Not without seeing all of your code. If you want to show it to me, zip up just the source code that is required to reproduce your simulation (no dlls please) and send it to ted dot carnevale at yale dot edu
Post Reply