www.neuron.yale.edu

I am trying to build a calcium dependent potassium channel by writing a mod file, since I don't think I can incorporate calcium dependence in the channel builder.
However, the intracellular calcium concentration that controls the potassium current changes based on 2 other calcium channels.

Where

tauca*(d[ca]i/dt) = -f(Icat + Icas) - [ca] + [ca]0;
where f, tauca and [ca]0 are constants 14.96uM/nA, 200ms, and 0.05uM respectively.

Icat and Icas are hh channels defined w/ minf, hinf, and their respective taus.
I have built those channels using the channel builder.

I'm a beginner at writing mod files, and I have looked at some examples, but I'm not sure where, and how I would be able to incorporate the intracllular calcium concentraion and its governing equation into the mod file where the mod file can read in the Icat and Icas that I built from channel builder.

Thanks in advance!

You can read and write (the latter with limitations) ion concentrations. In this recent post you will find towards the bottom some usefull tips, explanations and references:

https://www.neuron.yale.edu/phpBB2/viewtopic.php?t=492

I'm sorry, I just read the thread you refered to, but I still can't seem to understand how I am suppose to incorporate the currents from the other channels Icat and Icas that I have built.
I don't know how I should refer to those currents in my equations in the mod file, since I have built those channels simply using channel builder.
If my understanding is correct, nrngui will compile the mod file before anything else, and I wouldn't have even loaded my ses file in gui. How would gui know what I was referring to in the mod file then? On top of that, I wouldn't know what is the correct "code" or symbol that I should be using to refer to those channel builder built channels...

Maybe I'm just not seeing something in that thread, or not understanding it.

point me to the light.

Thanks.

Well you have no direct access to those currents, but I assumed you want to have access to the effect they have. Their effect is (next to membrane potential change) the change in calcium concentration. You do have access to the calcium concentration from your mod code.

The remark

Ted wrote: Multiple mechanisms can assert
USEION x . . . WRITE ix
(this ensures conservation of mass when there are multiple sources of
transmembrane ionic flux). However, only one mechanism is allowed to
WRITE the corresponding concentration xi or xo.

directly addresses this.

In the NMODL documentation and in the Neuron book you should be able to find this discussed in more detail.

so you are saying since the potassium channel depends on the calcium concentration and I have access to the calcium concentration, I should just directly use those calcium concentration in the hh equations that define this channel. Instead of using the equation that I stated above of how the calcium concentration are related to those other 2 currents, NEURON would already have calculated the calcium concentrion once I insert the 2 calcium channel in it.

am I understanding you correctly?

Thanks a lot, I think i know what you are saying now.

pzhang wrote:I am trying to build a calcium dependent potassium channel by writing a mod file, since I don't think I can incorporate calcium dependence in the channel builder.

Actually you can, for transitions between states. On the Properties page
of the ChannelBuilderGateGUI tool, click on the transition, then click on
EquationType / alpha,beta / Ligand
and select cai (or cao, if appropriate) if it is in the list. If it is not in the
list, select
New Ligand
and then enter
ca 2
in the edit field of the panel that pops up (unless
ca 2
is already there, in which case just click the Accept button).

However, the intracellular calcium concentration that controls the potassium current changes based on 2 other calcium channels.

So you need an "accumulation mechanism" for cai.

Icat and Icas are hh channels defined w/ minf, hinf, and their respective taus.
I have built those channels using the channel builder.

So you used the Channel Builder's
Properties / Selective for ion / ca
to assert that Icat and Icas produce ca currents, right? If so, you only
need a mod file for a mechanism that contains a
USEION ca READ ica, cai WRITE cai
or, if it includes a ca pump that is sensitive to extracellular ca
concentration,
USEION ca READ ica, cai, cao WRITE cai
These USEION statements allow the ca accumulation mechanism to
discover _all_ currents produced by mechanisms that WRITE ica
(including "ca selective" currents that you created with the Channel
Builder).

I'm a beginner at writing mod files, and I have looked at some examples, but I'm not sure where, and how I would be able to incorporate the intracllular calcium concentraion and its governing equation into the mod file where the mod file can read in the Icat and Icas that I built from channel builder.

In
C:\nrn59\examples\nrniv\nmodl
you'll find
cacum.mod very simple calcium accumulation in a shell next to the membrane
capump.mod accumulation with radial diffusion (concentric shells) and a pump
cabpump.mod adds "fast buffering" (actually instantaneous)
cadif.mod and cadifusl.mod involve "non instantaneous" buffering plus radial and longitudinal diffusion

Yes, you understood it correctly.

If in a mod-file you use the USEION statement in the NEURON block: you hook into NEURON's internal administration of ion-concentrations and currents.

With the READ part you get access to the ion-concentrations and ion-currents, which you can then use in your mechanism. Because the calcium dynamics is not limited to the ion-channels you inserted, you will find that just keeping track of those currents will probably not give a reasonable result. The simplest remedy is to include a calcium pump which gives exponential decay towards a specified target calcium concentration if there is no calcium influx. More advanced solutions also model calcium diffusion within the cell. (See the examples Ted pointed to.)

To give you another example of an implementation of calcium dependent potassium channel and a calcium pump you can have a look at the mod-files for the paper Mainen ZF, Sejnowski TJ (1996) 'Influence of dendritic structure on firing pattern in model neocortical neurons.' Nature 382:363-6 in ModelDB:
http://senselab.med.yale.edu/senselab/m ... model=2488
The model is ready to run so you can see it in action by just pressing the Auto-launch button.