From p.gleeson at ucl.ac.uk  Thu Sep 13 15:00:27 2007
From: p.gleeson at ucl.ac.uk (Padraig Gleeson)
Date: Thu, 13 Sep 2007 15:00:27 +0100
Subject: [neuroConstruct] [Fwd: neuroConstruct User Group]
Message-ID: <46E9427B.8020907@ucl.ac.uk>

Hi Derek,

Replying on the neuroConstruct mailing list. If there are any other 
questions which may be of general interest to users, send them to this 
address and I'll try to handle them.

MorphML files can be imported into neuroConstruct via the tab Cell 
Types. Select Add new cell type to project, and in the drop down box 
select NeuroMLConverter. Specify where the file is and press Create. It 
might be best to import first into a new project (containing the example 
cells, channels, etc.) to test.

The imported cell will probably be invalid if there are no channels 
specified (i.e. if it's not Level 2). Viewing the cell and placing the 
LeakConductance on all sections is a good start towards a valid cell in 
this case (default values of membrane leak conductance, capacitance, 
axial resistance will have to be updated though). Even if the MorphML 
file is Level 2 (strictly then a NeuroML file aot MorphML) and channels 
are present, these might not necessarily be in the project (e.g. if the 
cell has been exported from NEURON with the files at: 
http://neuroml.svn.sourceforge.net/viewvc/neuroml/nrn2NeuroML/). These 
channel names can be made match the cell mechanism names used in the nC 
project by a find & replace on the XML file before import, though I'm 
working to automate this when the file is imported.

MorphML/NeuroML files from any source can be imported (as long as they 
are valid at http://www.morphml.org:8080/NeuroMLValidator), so if there 
is any problem with import of a self generated, valid file, please let 
me know. Also, manually editing of a MorphML file and reimporting it 
might be the best way to edit cells if there is a specific function not 
implemented through the neuroConstruct GUI (e.g. bulk renaming of 
sections/groups).

Hope this helps,
Padraig






-------- Original Message --------
Subject: 	neuroConstruct User Group
Date: 	Thu, 6 Sep 2007 10:05:57 -0500
From: 	Derek James <djames at gmail.com>
To: 	p.gleeson at ucl.ac.uk



...

I'm interested in modeling the neocortex, at a fairly coarse-grained 
level, and I'm planning on generating XML files to be imported into 
neuroConstruct for visualization. The tutorial simply says "morphML 
files may be imported into neuroConstruct", but isn't very specific 
about how this is done.

...

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.mailinglists.ucl.ac.uk/pipermail/neuroconstruct/attachments/20070913/d897f1af/attachment.htm>