Biophysic hierarchy in the GUI

[hallockk]

I have an internodal length that is divided into 10 individual sections. The entire internode is also represented as a subsection. When I change the parameters of the internodal subsection, the parameters of the individual sections do not reflect that change.

Is there a way to have the GUI reflect those changes?

Thanks,

Kevin

[ted]

I don't understand the question.

NEURON has sections and segments. What is a "subsection"?

I have an internodal length that is divided into 10 individual sections. The entire internode is also represented as a subsection.

If an internode is divided into 10 sections, how can it also be represented as a "subsection"?

When I change the parameters of the internodal subsection, the parameters of the individual sections do not reflect that change.

I can't understand this until I know what a "subsection" is. Also, I need to know how are you changing the parameters. With assignment statments in hoc? or by changing the value in a numerical field of a GUI tool?

Is there a way to have the GUI reflect those changes?

Are you asking about a particular GUI tool? Do you want the appearance of a "stick figure diagram" of a model cell to change, or do you want the numerical value displayed in some numerical field of a particular GUI tool to change?

[hallockk]

I have an internode that has 10 segments and all of the segments are in one section. I'd like to be able to change a parameter for the section in the GUI and have it reflected in the values of each segment in the section. I could change it using hoc, but I don't think those changes aren't reflected in the GUI.

[ted]

I'd like to be able to change a parameter for the section in the GUI and have it reflected in the values of each segment in the section.

I just did this, with a model cell managed by a CellBuilder, and that's exactly what happened. What GUI tool are you using to change parameters?

I could change it using hoc, but I don't think those changes aren't reflected in the GUI.

True and true.

an internode that has 10 segments

An even number for nseg. That's OK as long as you never have any code that refers to a range variable at 0.5, e.g. statements with terms similar to
dend(0.5)
or its synonym
dend
Have you read
Why should I use an odd value for nseg?
on NEURON's FAQ page
http://www.neuron.yale.edu/neuron/faq/general-questions
?

[hallockk]

My mistake. I meant ten sections, not segments. I have the following ten sections that represent one internode:

node-internode_section-internode_section[1]-internode_section[2]-internode_section[3]-internode_section[4]-internode_section[5]-internode_section[6]-internode_section[7]-internode_section[8]-internode_section[9]-node[1]-

I included the nodal sections on either side as well.

Those ten sections are in one subset (Internode00). When I change a biophysical parameter of the subset, the values of the each section remains unchanged in the GUI.

Is there a way to have a change in the subset extend to each section in within the subset while still maintaining the ability to change each section individually?

Sorry about my confusion.

Thanks.

[ted]

You didn't answer my question

Code: Select all

What GUI tool are you using to change parameters?

but you're talking about subsets, so I will presume that you are using the CellBuilder.

Code: Select all

Is there a way to have a change in the subset extend to each section in within the subset while still maintaining the ability to change each section individually?

With "Specify Strategy" checked, you could select the subset and also each of the individual members of the subset.

For example, suppose I had a subset called main that contained soma, dend, and axon. On the Biophysics page, with "Specify Strategy" checked, the middle column of the CellBuilder would contain these names:
all
main
soma
dend
axon

Then when I clear the "Specify Strategy" checkbox, the Biophysics page would allow me to set the properties of all, main, soma, dend and axon. Since the model specification is executed from top to bottom, anything I do to all could be overridden by whatever I do to main. Likewise, anything I do to main could be overridden by whatever I do to the individual sections soma, dend, or axon.

This is OK for simple models, but it becomes inconvenient if the model is even a little complex. I find that the CellBuilder is great for dealing with subsets, but as soon as I need to deal with individual sections, I have to turn off the CellBuilder's Continuous Create button (this can be done with a hoc statement) and start using hoc code that I write myself.

I often find it useful to do something like this:

Code: Select all

load_file("nrngui.hoc")
load_file("cell.ses") // a CellBuilder saved with Continuous Create on
CellBuild[0].continuous = 0 // turn off Continuous create
  // to keep the CellBuilder from overwriting any changes that I make

proc chage_some_section() {
  . . . statements that tweak the parameters of one or more sections . . .
}

change_some_section()

[hallockk]

This is OK for simple models, but it becomes inconvenient if the model is even a little complex. I find that the CellBuilder is great for dealing with subsets, but as soon as I need to deal with individual sections, I have to turn off the CellBuilder's Continuous Create button (this can be done with a hoc statement) and start using hoc code that I write myself.

I was hoping I could do it from within the GUI, but I guess coding is the way to go.

Thanks.