Provide uncurated data sets from BBP in open data format?


#1

Hello,

my name is Andrej Warkentin and I’m a master student at the Bernstein
Center for Computational Neuroscience Berlin
(https://www.bccn-berlin.de/Graduate+Programs/1_Master+Program/Current_Students/).

I would like to fit models to the uncurated data set from the BlueBrain Project
(http://microcircuits.epfl.ch/#/uncurated). My problem is, that I can’t
open all the IGOR Pro files (including the meta information) because I
don’t have that Software. Would it be possible to provide the data in an
open data format like hdf5 or ideally in the NeurodataWithoutBorders
(http://www.nwb.org/) format ?
Or if that’s not possible at least a script (ideally in Python but Matlab would also be okay)
that extracts all the information from the files (I already have a script that can read the files
of the current and voltage traces but I’m also interested in the meta information).

Thanks very much and best regards,

Andrej Warkentin


#2

Hi Andrej,

The development version (master branch) of Neo supports reading IGOR Pro files; Neo support for NWB is in development - see https://github.com/apdavison/python-neo/tree/nwb and https://github.com/NeuralEnsemble/python-neo/issues/221

I’ll try to find time to put together a Jupyter notebook demonstrating this.

Best regards,

Andrew


#3

Here is a Jupyter notebook

https://collab.humanbrainproject.eu/#/collab/1612/nav/14273

which downloads a data file from the BBP portal, imports one of the Igor Pro files, and exports an NWB file. The NWB file produces a couple of validation errors (as I said, NWB support is still under development), but this could be a good starting point for you.

Andrew


#4

Yep, that definitely helps already. Thank you !

Do you happen to know if for the uncurated data there is also the cell gene expression information somewhere to find ? I’ll have a look at it myself but if you already know it that would make it easier of course. Or maybe you know who would know it ?

Cheers,
Andrej