non-uniform distribution of channels

[thats_karlo]

Dear Friends,

As you know the parameter "nseg" devide a section with lenght "L" into
(L/nseg) equal segment. and NEURON computes the time course of variables e.g. mebrane potential at the center of this segment.
if we assume the the density of channles are unifrom over the surface, then we choose "nseg=1"

as i undrestand it means

for uniform channel distribution, and "nseg=4" we devide a section to 4 equal part and
----------------------------------
|V1(t) |V2(t) |V3(t) |V4(t) |

V1(t)=V2(t)=V3(t)=V4(t) (thus one segment was enough)



but when we have diffrent channel distribution, time course of membrane potential at this four segment are Not equal.(Am i right?)

if yes, i have 3 questions:

1- How can we attribute a special channel density,e.g. diffrent G_max for Na+, K+ or Ca2+ for every part of section?

2- for example, i like to have a Guassian distribution of Na+ channel over the the surface of the section? what should i do?

3- How can we plot the V1(t),V2(t),.....?


Thank's in advance

[ted]

if we assume the the density of channles are unifrom over the surface, then we choose "nseg=1"

Not really. Here's the right way to think about this:
If the membrane current density is sufficienty uniform that it can be accurately
represented by the value at the midpoint of the section, then nseg can be 1.
If not, you need to chop space into smaller bits (i.e. use a larger value for nseg).
Notice that "membrane current density is sufficiently uniform" is not the same as
saying that "membrane current density is uniform," nor is it the same as saying that
"ion channel density is uniform." A very efficient strategy for getting good spatial
accuracy is described in chapter 5 of The NEURON Book. You can also read about
it in
Hines, M.L. and Carnevale, N.T. NEURON: a tool for neuroscientists.
The Neuroscientist 7:123-135, 2001.
which you can get by going to NEURON's home page
www.neuron.yale.edu
and then clicking on
Papers about NEURON

"nseg=4" we devide a section to 4 equal part and
----------------------------------
|V1(t) |V2(t) |V3(t) |V4(t) |

It is not a good idea to use an even value for nseg, because it makes the value of
v and other range variables at the midpoint (0.5) ambiguous. For your example with
nseg = 4, what should hoc return when you ask for v(0.5)? the value at 0.375, or the
value at 0.675? The value you get will be determined by roundoff error.

1- How can we attribute a special channel density,e.g. diffrent G_max for Na+, K+ or Ca2+ for every part of section?

2- for example, i like to have a Guassian distribution of Na+ channel over the the surface of the section? what should i do?

3- How can we plot the V1(t),V2(t),.....?

These are all variations on the question "how to access a range variable." Range
and range variables are discussed in chapter 5 of The NEURON Book. You will also
find useful information on these topics in
Hines, M.L. and Carnevale, N.T. The NEURON simulation environment.
Neural Computation 9:1179-1209, 1997
which is downloadable from the same page as "NEURON: a tool for neuroscientists."

--Ted