What does Init do?

[ecchiang]

Hello,

I've just started using Neuron after 2 years of patch clamping in brain slices. When I patch clamp on the rig in Current clamp, I can add positive/negative current to maintain a resting membrane potential, usually -70 mV. I thought that setting the value "Init (mV)" in the RunControl panel would be similar, but it does not seem to be the case. At -65 mV, the soma.v(0.5) will remain constant but at any other value, the voltage will change as soon as the trace starts (no mechanism inserted). What does this mean?

For example, let's say I insert the code:

create soma
access soma
insert hh
soma.nseg = 1

Then I open a voltage graphc (graphing v(.5))

From RunControl,
if Init(mV) = -65, its a straight line
if Init(mv) = -90, there's an AP
if Init(mV) = -40, the voltage hyperpolarizes to almost -80 very quickly, than slowly depolarizes

However, typically in cells that I've recorded from, at -40 mV you might get spontanous APs, and at -90 mV, I have never gotten a spontanous AP.

I realize that real cells in the brain have more channels than just hh. I'm just trying to understand what init(mV) really sets, and what happens afterwards in Neuron.

Thanks in advance for any insight.

Elizabeth
[/code]

[ted]

When I patch clamp on the rig in Current clamp, I can add positive/negative current to maintain a resting membrane potential, usually -70 mV.

This is a "steady state initialization under current clamp."

I thought that setting the value "Init (mV)" in the RunControl panel would be similar, but it does not seem to be the case.

Clicking on the Init button does a "steady state initialization under voltage clamp." That
is, membrane potential is set to the value displayed in the numeric field next to the Init
button, then all voltage-dependent states are set to their steady state values for that
membrane potential.

Clicking on the Init & Run button does a "steady state initialization under voltage clamp",
and then lets the model run free. It's as if you had clamped the entire cell--all of it, not
just the place where your patch electrode is located--to the displayed membrane
potential, and let it sit there for a minute or two. Then, when you start your oscilloscope
sweep, you release the cell from voltage clamp and just let it do whatever it's going to do.

At -65 mV, the soma.v(0.5) will remain constant but at any other value, the voltage will change as soon as the trace starts (no mechanism inserted).

Are you quite sure about the "no mechanism inserted" bit? Your code

Code: Select all

create soma
access soma
insert hh

sure as heck inserts the hh mechanism.

From RunControl,
if Init(mV) = -65, its a straight line
if Init(mv) = -90, there's an AP
if Init(mV) = -40, the voltage hyperpolarizes to almost -80 very quickly, than slowly depolarizes

However, typically in cells that I've recorded from, at -40 mV you might get spontanous APs, and at -90 mV, I have never gotten a spontanous AP.

Let's analyze these situations, one at a time.

hh likes to rest at -65 mV, so you wouldn't be surprised that, if you voltage clamp it to
-65, then release it, it will stay put at -65 mV.

What if you voltage clamp a cell to -90 mV for a minute or two, then release the clamp?
A lot of cells would show anode break excitation. hh mechanism sure does.

What if you inject depolarizing current that drives membrane potential to -40 mV, and
hold it up there for a few minutes? The hh mechanism's gna inactivates, and gk turns
on. Now shut off the depolarizing current. Would it be a surprise to see the cell zip down
to very near ek?

Initialization is discussed in some detail in chapter 9 of The NEURON Book. It even
shows you how to automate certain kinds of custom initializations, e.g.
how to let a model cell settle to its own steady state
how to guarantee that a model cell has a particular resting potential
how to save and restore the state of a model cell that is spontaneously active
how to initialize ionic concentrations in models that involve ion accumulation, transport,
buffering, and diffusion

[ecchiang]

Are you quite sure about the "no mechanism inserted" bit? Your code

You're right, maybe I meant a point process? (no IClamp, VClamp, SEClamp, alphasynapse, etc.)

This is a "steady state initialization under current clamp." ...

Clicking on the Init button does a "steady state initialization under voltage clamp."

Thanks, that explains a lot.

Initialization is discussed in some detail in chapter 9 of The NEURON Book.

Yup, I've actually read the whole book but I'm still trying to figure out what it means. :)

Elisa

[ted]

I've actually read the whole book

Thanks! I wondered why it had rocketed all the way up to #570,172 in Amazon's ranking of
book sales . . .