Public channel for all NeuroML related discussion. Documentation at https://docs.neuroml.org
OK, so it's compiling fine then. I expect we need to tweak something in pyneuroml for it to recognise this new mac format..
in the example, modify the
pynml.run_lems_with_jneuroml_neuron function call to skip compilation of mod files by adding the compile_mods=false argument to it. So it should use the files you just compiled and run correctly 🤞
I'll file us a bug about the pynml error, I think we also need to look for
libnrnmech.dylib to check for successful compilation
2 replies
See this error. When i run the export_nm2.py function in https://github.com/OpenSourceBrain/BahlEtAl2012_ReducedL5PyrCell/tree/master/NeuroML2 it results in https://paste.centos.org/view/ee91573c
1 reply
It fails to match with the actual simulation file on the repo.
When we reinstalled neuron, it now correctly creates a
x86_64 directory with the compiled library.
in general, clean up your folders to remove these temporary files before each run. if you cloned using git, something like
git clean -dfx will do it for you
I did that too, but it also results in a mismatch with png file
3 replies
what's it supposed to look like? (I haven't run that model in a while)
shayan1618: will you tell me what commands you're running, and I'll see if I can replicate the issue here
First in /NEURON
nrngui exampl1.hoc then in /NeuroML2 python export_nml2.py
*example1.hoc
It works now. I have found the error. In _reducedmodel.hoc instead of
nrn_load_dll("./channels/arm_64/.libs/libnrnmech.so") it was supposed to be 'x86_64'.
shayan1618: it's x86_64 on the repo: https://github.com/OpenSourceBrain/BahlEtAl2012_ReducedL5PyrCell/blob/master/NEURON/reduced_model.hoc#L1
did you change that while debugging the earlier issues perhaps?
Yes, I made that change to make it run with the package that was compatible with my arch which was searching for the file
libnrnmech.so in the given "arm_64" dir.
it's tricky with arches etc---so keep an eye on these things.
best to get a working environment, and then keep it clean :)
I was trying to compare the channels created in NeuroML with the original NEURON files. Facing this error while running channel analysis on the .mod file. https://paste.centos.org/view/eb420eea
1 reply
I looked into the file HHAnalysis.py but was unable to find the ValueError output there. Here is the link to the mod file https://github.com/shayan823/BahlEtAl2012_ReducedL5PyrCell/blob/master/NEURON/channels/kfast.mod
It is running fine only for the nat channel, rest for all the other channels its throwing the same error.
may need Padraig to look into this: @pgleeson-54ac1a06db8155e6700e68c5 any ideas ^
@pgleeson: ^
(my client doesn't seem to tag people correctly)
@shayan1618:matrix.org The issue is that you need to have compiled the mod files in advance before you run the analysis routine. Run
nrnivmodl in the folder with the mod files, make sure the x86_64 folder is there and then run pynml-modchananalysis kfast -stepV 5 -temperature 37 -modFile kfast.mod. Worked fine for me.
@pgleeson: do we need to tweak the script to make it check for the compiled mod file folders maybe (or make it compile them as part of the run if possible)?
1 reply
I am finding trouble in converting kslow channel i.e. https://github.com/shayan823/BahlEtAl2012_ReducedL5PyrCell/blob/master/NEURON/channels/kslow.mod , it has a fractional gate. I searched for examples but was unable to find one. Any example on that will be very helpful for now. Ankur Sinha (FranciscoD@fedora) @pgleeson
1 reply
Unable to validate the kslow nml file though i.e. https://github.com/shayan823/BahlEtAl2012_ReducedL5PyrCell/blob/master/NeuroML2/kslow.channel.nml. It is throwing error https://paste.centos.org/view/05461d29. Any idea? Rest all the channel files are validated.
Thanks @shayan1618:matrix.org. I'll look into that, but I'm off work soon for a couple of days. As a general point though, Gitter is fine for general questions, but if it is a specific issue related to a repo, a Github issue might be the bets place to ask, and then there's a record of it closer to the code, which can be resolved and closed (also useful if it does need to remain open). As discussed with Boris though, the fractional gate may not have been fully tested and I suspect the mapping to use it on Neuron may need some work. I can look into this next week. Your work so far on the Bahl model has been very valuable though.
Understood. I will make sure of following on the github page for the specified issues.
I don't think we've advertised HARMONY 2022 here folks! Please register, it's free:
A reminder about the NeuroML Development Workshop later today!
Time: 15:00-18:00 UTC. Scheduled for 3 hours, but everyone's busy, so let's try to get through the agenda in 2 hours, and leave the rest for open discussion.
Zoom link: https://ucl.zoom.us/j/91720715131?pwd=YmtLa29jd0ROWnNMVGovVzhCd0NIdz09 (please register for Harmony too!)
Agenda: https://docs.google.com/document/d/1GOkmylIlLH3dbxB2Pmis7VxFPzipstLBdXn5uBu4rv8/edit
I'm trying to write some NeuroML to capture arbitrary kinetics that I can then convert via pynml into a MOD file for simulation, and I'm getting close to understanding how to do it, but I'm not quite there... what I have is: https://www.dropbox.com/s/c1oqm6dvsd8u1ex/test6.nml?dl=0
Questions:
- I thought putting species="k" would cause it to generate an "ik" not a nonspecific current... how do I do this?
- It looks like the NeuronWriter only allows using one species at a time, with the possible exception of things that are calcium dependent. Is that true? If not, is there a way to have mgi here -- in our "potassium" channel model -- be what you'd get from "USEION mg READ mgi"
- Conversion automatically includes an assigned of "celsius" in the generated MOD file; is this variable available in all exported formats or how can I write the nml such that it'll translate correctly to mod but also work on other platforms?
- What's the intended role of the "conductance" parameter. Both it and gmax are automatically added to the MOD file, but the current calculation is forced to be ohmic (or ghk, or maybe a couple other fixed formats)... and that gion seems hard-coded in the NeuronWriter to be based on gmax not conductance?
- Any gotchas I should be aware of with the NeuronWriter? I'm mostly basing my understanding of how to write NeuroML by trying things and seeing what happens when I translate to MOD files
1 reply
I've asked it on the GH repo now too, so that it can be answered/discussed there: NeuroML/NeuroML2#187
Ankur Sinha (FranciscoD@fedora): What's the difference in the file libnrnmech.dylib and libnrnmech.so files created once we compile the mod files for the density mechanisms?
1 reply
I was unable to locate the scripts for
pynml-modchananalysis and pynml-channelanalysis . Help me locate them.
1 reply
that way, all the bits are part of the API, and a "main" function allows us to also provide commands
Hi! I am currently trying to convert the all-active cell model 497233223 of this electrophysiology.
The cell contains information about e_pas and g_pas which are the leakage reverse potential and specific leakage conductance of the pas channel respectively.
When converting this to NeuroML, the e_pas value is set using the erev parameter (see here).
How to translate g_pas to NeuroML? @pgleeson @sanjayankur31