monitor the attenuation of the peak value of the action potential with distance from the ion channel

The basics of how to develop, test, and use models.
Post Reply
Sun Xiaoqing
Posts: 32
Joined: Wed Aug 28, 2019 1:18 am

monitor the attenuation of the peak value of the action potential with distance from the ion channel

Post by Sun Xiaoqing » Sat Sep 28, 2019 9:15 am

Good day,

I'm implementing a simple unmyelinated axon, I want to use LFPsim method to measure the voltage potential at some different compartments. My purpose is to monitor the attenuation of the peak value of the action potential with distance from the ion channel.

My question is:
1. How can I apply a stimulus at a specific ion channel (Has to be ion channel current, originated from concentration gradient and electric potential
gradient) to make the neuron fire an action potential.
2. How to extract the peak value of the action potential at different positions?
3. I want to polt the peak value of action potential(y-axis) and is corresponding L(the distance between the ion channel
which is applied with stimulus). Is NEURON capable of doing that? If not, how can I save the value that I measured
and plot it using other software like MATLAB?

The hoc code of the cell is:

Code: Select all

proc celldef() {
  topol()
  subsets()
  geom()
  biophys()
  geom_nseg()
}

create soma, axon

proc topol() { local i
  connect axon(0), soma(1)
  basic_shape()
}
proc basic_shape() {
  soma {pt3dclear() pt3dadd(0, 0, 0, 1) pt3dadd(15, 0, 0, 1)}
  axon {pt3dclear() pt3dadd(15, 0, 0, 1) pt3dadd(90, 0, 0, 1)}
}

objref all
proc subsets() { local i
  objref all
  all = new SectionList()
    soma all.append()
    axon all.append()

}
proc geom() {
  soma {  L = 30  diam = 30  }
  axon {  L = 500  diam = 1  }
}
proc geom_nseg() {
   soma { nseg = 1  }
   axon { nseg = 10  }
}
proc biophys() {
  forsec all {
    Ra = 100
    cm = 1
  }
  soma {
    insert hh
      gnabar_hh = 0.12
      gkbar_hh = 0.036
      gl_hh = 0.0003
      el_hh = -54.3
  }
  axon {
    insert pas
      g_pas = 0.002
      e_pas = -65
  }
}
access soma

celldef()
And I save the point process and run control to another session file. The result that I plot seemed fine but the initializing an action potential by a pulse of an Iclamp at soma wasn't the condition that I wanted.

Another question is , after double-clicking axon.hoc I loaded the session file and error messages were shown:

Code: Select all

NEURON -- VERSION 7.7.2 7.7 (2b7985ba) 2019-06-20
Duke, Yale, and the BlueBrain Project -- Copyright 1984-2018
See http://neuron.yale.edu/neuron/credits

oc>Window menu not closed
C:\nrn\bin\nrniv.exe: A menu is still open
 in windows_of_axon.ses near line 194
 }
  ^
        xpanel(0, 24)
      xopen("windows_of...")
    NEURONMainMenu[0].execute1("{xopen("wi...")
  NEURONMainMenu[0].load_file(1, "C:/models/...")
and others
The final problem is I used the LFPsim from ModelDB:
https://senselab.med.yale.edu/ModelDB/S ... 140#tabs-1

I added the file folder of the LFPsim and put flp.mod and mea.mod in the same mod folder under my axon file. I have successfully compiled the mod file, but when I typed at the nrniv window, I got an error:

Code: Select all

xopen("extracellular_electrode.hoc")
C:\nrn\bin\nrniv.exe: Can't open  extracellular_electrode.hoc
 in axon.ses near line 57
 xopen("extracellular_electrode.hoc")
                                     ^
        xopen("extracellu...")
I didn't understand what was the reason of that.

Thanks a lot for your time, hope that anyone could give me some suggestions.

Sun Xiaoqing
Posts: 32
Joined: Wed Aug 28, 2019 1:18 am

Re: monitor the attenuation of the peak value of the action potential with distance from the ion channel

Post by Sun Xiaoqing » Sat Sep 28, 2019 11:19 am

I searched the ModelDB, but I didn't find suitable AP initialize mechanism(change ion concentration or voltage gradient at ion channel), maybe my searching was not enough. If anyone knows some examples that I can reference to, please tell me. Thanks in advance.

Sun Xiaoqing
Posts: 32
Joined: Wed Aug 28, 2019 1:18 am

Re: monitor the attenuation of the peak value of the action potential with distance from the ion channel

Post by Sun Xiaoqing » Sat Sep 28, 2019 11:41 am

I encountered this model which I thought I can use the mod file.:
https://senselab.med.yale.edu/ModelDB/S ... tml#tabs-2

ted
Site Admin
Posts: 5633
Joined: Wed May 18, 2005 4:50 pm
Location: Yale University School of Medicine
Contact:

Re: monitor the attenuation of the peak value of the action potential with distance from the ion channel

Post by ted » Sat Sep 28, 2019 9:48 pm

It sounds like you're starting a modeling project and need to work closely with a collaborator on it for two reasons. First, to define exactly what you are thinking about. A big problem is that this sentence doesn't make sense:
"How can I apply a stimulus at a specific ion channel (Has to be ion channel current, originated from concentration gradient and electric potential
gradient) to make the neuron fire an action potential."
And second, to make sure that what you end up with makes sense to you. This goes beyond the aims of the NEURON Forum, which is to address focussed questions on how to use NEURON.

Sun Xiaoqing
Posts: 32
Joined: Wed Aug 28, 2019 1:18 am

Re: monitor the attenuation of the peak value of the action potential with distance from the ion channel

Post by Sun Xiaoqing » Sun Sep 29, 2019 7:34 am

Thank you for your reply!

Sorry for my unclear statements and purposes. What I'm trying to do is first I need to construct a multi-compartment HH model and then use LFPsim to measure the extracellular potential along the axon. I want to see the extracellular potential generated by the transmembrane current which can be completed using LFPsim. Does that make sense now? Apologize again for my previous statement and thank you for pointing that out.

ted
Site Admin
Posts: 5633
Joined: Wed May 18, 2005 4:50 pm
Location: Yale University School of Medicine
Contact:

Re: monitor the attenuation of the peak value of the action potential with distance from the ion channel

Post by ted » Sun Sep 29, 2019 1:34 pm

That's much clearer, thanks. At a minimum, you need a model axon, and something to stimulate the axon. You should also have a graph that shows membrane potential along the length of the axon for at least the following reasons:
1. To verify that the stimulus is big enough to trigger a spike but not so big that it produces an unacceptably large stimulus artifact. Basically you want a stimulus current that is very short in duration (0.1 ms will do) so that desensitization during the stimulus is minimal, and has an amplitude that is about 2 x threshold. You'll have to discover the threshold current amplitude for yourself, because it depends on the anatomical and biophysical properties of your axon model, and the assumed temperature.
2. To verify that the axon is long enough that spike propagation actually happens. You want the axon to be at least twice as long as what might be called the "wavelength" of the action potential. By "wavelength" I mean a length of axon that is sufficiently long to capture the "entire" action potential waveform. If there is a significant post-spike hyperpolarization, the "wavelength" is roughly the distance between the advancing front of the action potential (the point at which membrane potential has risen by a few mV above resting potential) and the end of the post-spike hyperpolarization. If there is no post-spike hyperpolarization, it's just the distance between the advancing front of the spike and the point at which membrane potential has fallen to within a few mV of resting potential.
3. To verify that the axon is long enough that there is a significant stretch of axon that is unaffected by sealed-end effect. Never heard of sealed-end effect? Read
Recording membrane voltage along an axon
viewtopic.php?t=3167#p13158

I almost forgot to mention that you'll want to be sure that spatial discretization of your model axon is sufficient for spatial accuracy. The easy way to do that is build the model axon with the CellBuilder and use the d_lambda rule to ensure that no compartment is longer than 0.1 times the AC length constant at 100 Hz.

Sun Xiaoqing
Posts: 32
Joined: Wed Aug 28, 2019 1:18 am

Re: monitor the attenuation of the peak value of the action potential with distance from the ion channel

Post by Sun Xiaoqing » Mon Sep 30, 2019 7:53 am

Dear Ted,

I truly appreciate your timely help, I really learned a lot from your expertise reply. I've successfully built my model (one with axon implemented with hh mechanism, the other one with passive conduction).

My next step is to First, construct a model with soma(hh), axon(pas). Next, apply a threshold stimulus at the soma to invoke an action potential along the axon. Then, find out the position of the axon (x) where the AP decreases to 50%. Finally, plot the x with axons of different diameters.

For the model: soma(hh), axon(hh), plot the propagation velocity with different diameters.

Could you give me some suggestions? Or any post I can refer to? Again thank you so much for your patient reply.
Last edited by Sun Xiaoqing on Tue Oct 01, 2019 10:14 am, edited 6 times in total.

Sun Xiaoqing
Posts: 32
Joined: Wed Aug 28, 2019 1:18 am

Re: monitor the attenuation of the peak value of the action potential with distance from the ion channel

Post by Sun Xiaoqing » Tue Oct 01, 2019 10:12 am

I'm reading this post which I find similar to my case:viewtopic.php?f=15&t=1483
but I have confusion about how to get my half maximum vale of AP during the hyperpolarizing process.
and how to get (x vs Diameter of the axon) and make a plot of it.

ted
Site Admin
Posts: 5633
Joined: Wed May 18, 2005 4:50 pm
Location: Yale University School of Medicine
Contact:

Re: monitor the attenuation of the peak value of the action potential with distance from the ion channel

Post by ted » Tue Oct 01, 2019 12:49 pm

If your primary interest is in spike propagation along an axon, why include a soma in your model? Physiologists who study spike propagation along sciatic nerve or squid axon almost never pay any attention to the soma, and typically work on isolated axon without a soma.

If your interest is in how spike amplitude varies as it propagates along an axon, why bother with extracellular recording?
how to get my half maximum vale of AP during the hyperpolarizing process
You have a lot of curiosity, which is good. You'd probably benefit greatly from an introductory course in neurophysiology, with a laboratory component. From that you would know that spike amplitude is typically measured as [v at depolarized peak of the spike] - resting potential. I have never encountered the term "half maximum value of AP" in experimental or computational neuroscience, and have no idea what you mean by it or why it would be useful.

Sun Xiaoqing
Posts: 32
Joined: Wed Aug 28, 2019 1:18 am

Re: monitor the attenuation of the peak value of the action potential with distance from the ion channel

Post by Sun Xiaoqing » Wed Oct 02, 2019 9:28 am

Dear Ted,

Thank you for introducing me to the introductory course in neurophysiology, although I'm very interested in that course, unfortunately, I'm still in my third year of university and it's such a regret that I can't make it.

During the process of discussing with my lecturer, new ideas came out and I might have to change the method of simulations. I apologize for the inconvenience that I caused in this thread, and I'll open another topic to discuss my current problem. Thank you for replies to my confusion.

Best regard,
XIAOQING

Post Reply