NEURON: section stack overflow

A Python package that facilitates development and use of models of biological neural networks

Moderator: tom_morse

Post Reply
Mohamed_Hisham
Posts: 16
Joined: Thu Oct 08, 2015 2:56 pm
Location: Wright State University

NEURON: section stack overflow

Post by Mohamed_Hisham »

Hello.

I used to import cell template from a hoc file using the command:

Code: Select all

cellRule0 = netParams.importCellParams(label = 'FR',conds={'cellType': 'FR', 'cellModel': 'HH3D'}, fileName= 'FRcellTemplate.hoc', cellName='FR_Cell',importSynMechs=False)
netParams.cellParams['FRrule'] = cellRule0 	
the cell template contains the detailed 3D morphology of FR-Motoneuron with about 730 section.

the code was working fine on netpyne version 0.7.0 , but later tried to run my code on NSG ( neuroscience getway) and i got errors like:

Code: Select all

NEURON: section stack overflow
 near line 0
 ^
                                                                                                                                                                                                       FR_Cell[0].den5o3b3s1t2
                          
to make sure what is the problem, i updated my local PC netpyne version to 0.7.1 as well, and now I have the same Errors on my PC

Code: Select all

NEURON -- Release 7.4 (1370:16a7055d4a86) 2015-11-09
Duke, Yale, and the BlueBrain Project -- Copyright 1984-2015
See http://www.neuron.yale.edu/neuron/credits

loading membrane mechanisms from x86_64/.libs/libnrnmech.so
Additional mechanisms from files
 AXNODE.mod CaSm.mod IClampNoise.mod KdrIs.mod KdrSm.mod Llva.mod NafIs.mod NafSm.mod NapIs.mod RampIClamp.mod SinIClamp.mod TriangleIClamp.mod ax.mod dummy.mod h.mod info.mod xtra.mod
NEURON: section stack overflow
 near line 0
 ^
                                                                                                                                                                                                       FR_Cell[0].den5o3b3s1t2
                                                                                                                                                                                                     

NEURON: section stack overflow
 near line 0
 ^
*** longjmp causes uninitialized stack frame ***: python terminated

I would appreciate any help with this issue. Thanks in advance!
salvadord
Posts: 86
Joined: Tue Aug 18, 2015 3:49 pm

Re: NEURON: section stack overflow

Post by salvadord »

There were some bug fixes in the import code in v0.7.1 so it might have messed something. Could you please share the FRcellTemplate.hoc so I can try it out myself? thanks
salvadord
Posts: 86
Joined: Tue Aug 18, 2015 3:49 pm

Re: NEURON: section stack overflow

Post by salvadord »

For some reason when deleting sections after importing the cell they were accumulating on the stack, so had to use h.pop_section() to avoid error.

I have pushed the changes to the github "development" branch and tested your code successfully. I will release fix in next version.
Mohamed_Hisham
Posts: 16
Joined: Thu Oct 08, 2015 2:56 pm
Location: Wright State University

Re: NEURON: section stack overflow

Post by Mohamed_Hisham »

Thanks Salvador very much.

is there any modifications i can do manual within the NetPyNE 0.7.1 version so i can import the cell 3D structure. till the next version is released.
please let me know.

Thanks in advance
salvadord
Posts: 86
Joined: Tue Aug 18, 2015 3:49 pm

Re: NEURON: section stack overflow

Post by salvadord »

You can use the github development version (many of our lab members do it this way so they always have access to the latest features):

1) uninstall the pip version: pip remove netpyne
2) clone the github version: git clone https://github.com/Neurosim-lab/netpyne.git
3) make sure you are in the development branch: git checkout development and git pull
4) add the new netpyne path to your PYTHONPATH environment variable
Mohamed_Hisham
Posts: 16
Joined: Thu Oct 08, 2015 2:56 pm
Location: Wright State University

Re: NEURON: section stack overflow

Post by Mohamed_Hisham »

Thank you very much
Post Reply