Dear all,
I am very new to Neuron and I also am facing troubles using extracellular stimulation.
My goal is to evaluate the influence of various electrical stimuli (monophasic, biphasic, pseudo-monophasic... pulses) on a very simple model of an auditory nerve fiber. To do so I started studying the extracellular_stim_rec files that are available (thanks a lot for this Ted).
I first modified the cell to model a basic linear myelinated cell with unmyelinated nodes and soma.
I verified that the topology was correctly incorporated in Neuron and then modified the stimulus waveform.
Here again, the amplitude and time vectors for my stimulus seem to be correctly loaded but the intracellular voltage soma.v(0.5) doesn't make any sens for me.
It returns multiple spikes for about 20ms while my stimulus is only 0.4ms PLUS I still get these spikes even when the amplitude is set to zero !
It looks to me that I have another unwanted created somewhere in this code but I cannot figure out where.
Any help or suggestions would be great !
yours
Extracellular stimulation of a model cell
-
- Site Admin
- Posts: 6289
- Joined: Wed May 18, 2005 4:50 pm
- Location: Yale University School of Medicine
- Contact:
Re: Extracellular stimulation of a model cell
I split your post into a new discussion thread because diagnosing and eliminating your problem has little to do with the existing thread.
First question: what can you tell me about the behavior of your model cell itself? Did you implement and test it on its own, before trying to add the code that implements extracellular stimulation? That's the best time to diagnose and fix implementation problems with the model itself. The first task is to make sure that the representation of the biology--your own model--has been correctly implemented (that it is a close match to your conceptual model), and that it does what you think it should do, apart and aside from whatever it might do when a bunch of other code has been added to it.I first modified the cell to model a basic linear myelinated cell with unmyelinated nodes and soma.
Re: Extracellular stimulation of a model cell
Dear Ted, thank you so much for your reply.
To make sure I did everything the right way I started all again from the very beginning as in the example you propose for beginners. I implemented a simple cell (one soma + one dendrite both with HH behaviors) and it seems to be a more general issue:
When adding a classic IClamp I still get the same periodic spike train for soma.v(0.5).
So it doesn't have something to do with my cell definition or with the xtra mechanism.
When I have a closer look to v(0.5) I can see the contribution of the IClamp at the very beginning but then this unexpected spike train goes on for ever... (approx. 1 spike every 15 ms with a peak amplitude of 44.26 mV).
thanks again.
To make sure I did everything the right way I started all again from the very beginning as in the example you propose for beginners. I implemented a simple cell (one soma + one dendrite both with HH behaviors) and it seems to be a more general issue:
When adding a classic IClamp I still get the same periodic spike train for soma.v(0.5).
So it doesn't have something to do with my cell definition or with the xtra mechanism.
When I have a closer look to v(0.5) I can see the contribution of the IClamp at the very beginning but then this unexpected spike train goes on for ever... (approx. 1 spike every 15 ms with a peak amplitude of 44.26 mV).
thanks again.
-
- Site Admin
- Posts: 6289
- Joined: Wed May 18, 2005 4:50 pm
- Location: Yale University School of Medicine
- Contact:
Re: Extracellular stimulation of a model cell
Sounds like you have reproduced the problem in a simpler setting. That's extremely helpful for diagnosing and fixing it. Now for diagnosis.
Click on RunControl / Init.
Now bring up a Model View tool by clicking on NEURON Main Menu / Tools / Model View
What does the first line in Model View report?
Below that you should see these 4 lines:
* 1 real cells
0 artificial cells
0 NetCon objects
0 LinearMechanism objects
This is actually a browsable outline. Expandable items are marked by an *. Click on
* 1 real cells
Work through the outline and verify that everything it reveals is what it should be. Did this reveal any surprises or obvious problems?
Click on RunControl / Init.
Now bring up a Model View tool by clicking on NEURON Main Menu / Tools / Model View
What does the first line in Model View report?
Below that you should see these 4 lines:
* 1 real cells
0 artificial cells
0 NetCon objects
0 LinearMechanism objects
This is actually a browsable outline. Expandable items are marked by an *. Click on
* 1 real cells
Work through the outline and verify that everything it reveals is what it should be. Did this reveal any surprises or obvious problems?
Re: Extracellular stimulation of a model cell
Dear Ted, thanks for your help
I checked the Model View and found nothing surprising.
But I think I have identified several sources for my issues.
1/ I used values for gna_bar that were too low which I think are responsible for these periodic spikes.
2/ Once this problem solved I tried to implement a linear myelinated axon (myelinated sections + nodes + soma).
I found in another post that a first approximation for such a cell would be to use HH mechanism for the nodes and soma and passive mechanism for the myelinated sections.
When trying to run a simulations I notcied that I had clean spikes only when the electrode was located ON the soma, which no longer occurs when I use the same settings for different sections (HH for all).
If you have any suggestions on what parameters to use in such a model cell this would be very helpful. But I think I have no other choice but to read more about previous studies to find the proper parameters (eg. Try to simulate the model from Rattay_2001).
Thanks a lot.
Best
Quentin
I checked the Model View and found nothing surprising.
But I think I have identified several sources for my issues.
1/ I used values for gna_bar that were too low which I think are responsible for these periodic spikes.
2/ Once this problem solved I tried to implement a linear myelinated axon (myelinated sections + nodes + soma).
I found in another post that a first approximation for such a cell would be to use HH mechanism for the nodes and soma and passive mechanism for the myelinated sections.
When trying to run a simulations I notcied that I had clean spikes only when the electrode was located ON the soma, which no longer occurs when I use the same settings for different sections (HH for all).
If you have any suggestions on what parameters to use in such a model cell this would be very helpful. But I think I have no other choice but to read more about previous studies to find the proper parameters (eg. Try to simulate the model from Rattay_2001).
Thanks a lot.
Best
Quentin
-
- Site Admin
- Posts: 6289
- Joined: Wed May 18, 2005 4:50 pm
- Location: Yale University School of Medicine
- Contact:
Re: Extracellular stimulation of a model cell
I just added a published model of extracellular stimulation of a myelinated axon to ModelDB; I expect it will be made publicly accessible very quickly, and will post its URL when that happens. Examining it might help.
-
- Site Admin
- Posts: 6289
- Joined: Wed May 18, 2005 4:50 pm
- Location: Yale University School of Medicine
- Contact:
Re: Extracellular stimulation of a model cell
Model entry 239006
Extracellular stimulation of myelinated axon (Reilly 2016)
https://senselab.med.yale.edu/modeldb/S ... del=239006
Start by reading readme.html
Extracellular stimulation of myelinated axon (Reilly 2016)
https://senselab.med.yale.edu/modeldb/S ... del=239006
Start by reading readme.html
Re: Extracellular stimulation of a model cell
Thanks, I'll have a close look at it.
Best,
Quentin
Best,
Quentin