Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Create sample NeuroML connectome output #114

Closed
slarson opened this issue Jun 12, 2013 · 12 comments
Closed

Create sample NeuroML connectome output #114

slarson opened this issue Jun 12, 2013 · 12 comments

Comments

@slarson
Copy link
Member

@slarson slarson commented Jun 12, 2013

  • install NEURON and NeuroConstruct
  • load the connectome up in NEURON
  • set NEURON to write out data files for the voltage of all 302 neurons
  • graph the dataset using matplotlib to see all traces
  • upload graph and data files to GitHub
  • repeat data output, graphing and uploading steps for all compartments (302 x ~10)
@slarson slarson mentioned this issue Jun 12, 2013
0 of 3 tasks complete
@slarson
Copy link
Member Author

@slarson slarson commented Jun 19, 2013

@SuFizz good to talk to you today -- feel free to ask additional questions here on this thread

@Neurophile
Copy link

@Neurophile Neurophile commented Oct 3, 2013

Are you looking for something like this?
https://github.com/Neurophile/NEURONSimData
I have all the infrastructure to easily generate the full set. Graphs will need labels and such of course.

@pgleeson
Copy link
Member

@pgleeson pgleeson commented Oct 3, 2013

Thanks for setting that up.

Need to emphasize again: the neuroConstruct project, and hence the NEURON simulation it generated, does not yet produce electrically realistic behaviour. All the cells are passive (no voltage dependent ion channels) and the synapses are first guesses, based on vertebrate chemical synapses; invertebrates have more graded synaptic transmission.

So while getting NEURON to produce voltage traces is useful, the activity patterns are meaningless until we have more detailed cell & synapse models.

@Neurophile
Copy link

@Neurophile Neurophile commented Oct 3, 2013

There's not much point in generating the data/graphs at this time then.

I can, however, set up Python scripts to generate the graphs once we have a real model to make data for us. Is the desired output really a single graph with all 302 neuron voltage traces on it?

Also, either within neuroConstruct or as a standalone script, modify the CElegans.hoc file to output data for every segment in one run (hopefully that won't crash my laptop).

@pgleeson
Copy link
Member

@pgleeson pgleeson commented Oct 3, 2013

It's all useful stuff for when the model is more detailed.

Just managing expectations so that no one expects to extract any useful insights from the traces currently generated.

@Neurophile
Copy link

@Neurophile Neurophile commented Oct 4, 2013

Script for making graphs is up: https://github.com/Neurophile/NEURONSimData

Dump your neuroConstruct+NEURON sim output in /data, run the script and you get some basic graphs. More advanced visualizations to be added as needed/requested.

@slarson
Copy link
Member Author

@slarson slarson commented Oct 28, 2013

@Neurophile -- great job on getting this going. I have forked your repo back into the OpenWorm organization to raise its profile.

@Interintel had a nice way of graphing activity from his connectome engine project that would go beyond a couple of neurons. @Interintel can you point us to screen grabs that produced this so @Neurophile can see?

It is essentially a grid-based view where every cell is color mapped to show the activity level at a point in time. In order to add the temporal dimension, we would assemble these snapshots into successive frames and turn them into a movie.

Having the repo include .dat files taken from all 302 neurons to produce this graph would then serve up all the content needed to take the next step to get to #115, which I'll need to explain a bit more clearly.

@Neurophile are you still game to take the next steps on this?

@Interintel
Copy link

@Interintel Interintel commented Oct 28, 2013

Attached is a plot after running the Connectome with only the AVB, DB, DD,
VB and VD neurons ala Gait modulation in C. elegans: an integrated
neuromechanical model (2012) Jordan H. Boyle, Stefano Berri and Netta Cohen.

I haven't ran a full connectome plot for a while but will do this week.

Let me know what I can do and/or provide and I will do my best to get to
you,

Tim (@Interintel)

On Mon, Oct 28, 2013 at 1:53 PM, Stephen Larson notifications@github.comwrote:

@Neurophile https://github.com/Neurophile -- great job on getting this
going. I have forked your repo back into the OpenWorm organization to raise
its profile.

@Interintel https://github.com/Interintel had a nice way of graphing
activity from his connectome engine project that would go beyond a couple
of neurons. @Interintel https://github.com/Interintel can you point us
to screen grabs that produced this so @Neurophilehttps://github.com/Neurophilecan see?

It is essentially a grid-based view where every cell is color mapped to
show the activity level at a point in time. In order to add the temporal
dimension, we would assemble these snapshots into successive frames and
turn them into a movie.

Having the repo include .dat files taken from all 302 neurons to produce
this graph would then serve up all the content needed to take the next step
to get to #115 #115, which
I'll need to explain a bit more clearly.

@Neurophile https://github.com/Neurophile are you still game to take
the next steps on this?


Reply to this email directly or view it on GitHubhttps://github.com//issues/114#issuecomment-27255499
.

@slarson
Copy link
Member Author

@slarson slarson commented Oct 28, 2013

@Interintel whoops i think the attachment didn't come through. Maybe click into github and try dragging the image into your comment?

@Neurophile
Copy link

@Neurophile Neurophile commented Oct 29, 2013

I can do a heat map or Hinton diagram of the full connectome or whatever subset you care to generate. The graphs are populated automatically from whatever files are placed in the /data directory. The repo only has a couple of neurons simply because the data is for test only and I didn't want to upload a huge data set that people might mistake for "real" biologically realistic output.

I should also set up some command line options for generating different graph types or logical groupings of the data. 2D+time and 3D+time animations are also possible so don't be afraid to ask for whatever specialty visualization you want. If it is supported in matplotlib, you can have it.

@slarson
Copy link
Member Author

@slarson slarson commented Nov 2, 2013

Here's the attachment (attachments don't seem to work when replying to GitHub issue mails.

It is basically a heatmap. If we have some code that opens the dat files, builds a 4D heatmap, then we can easily go the next step to #115

Thanks both @Interintel and @Neurophile for the efforts on this!!

@slarson
Copy link
Member Author

@slarson slarson commented Dec 17, 2017

This all happens now within c302. Closing this.

@slarson slarson closed this Dec 17, 2017
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Projects
None yet
Linked pull requests

Successfully merging a pull request may close this issue.

None yet
4 participants