Exporting files for Neurodamus

[Hjorthmedh]

Missing information

• What file format should be used for the input spikes?

• When specifying synapse location, how can SEC_ID and SEC_X be derived from the morphology in the swc file? What is the algorithm that does the numbering of the sections? I have the coordinates of the synapses.

[Hjorthmedh]

Just to confirm, SEC_ID is the ID:s for a compartment assigned in the SWC file?

[Hjorthmedh]

Also received info from JD regarding input format:

1) spike format: the spike will be a single HDF5 sorted by gids

• data (dtype: float, shape: number_of_spikes)

• index_pointer(d:type: int, shape: number_of_gid + 1)

so you retrieve the spikes for a gid with:

data[index_pointer[gid]: index_pointer[gid+1].

However, my current implementation does not support this format yet. It supports a csv file with 2 columns, one for spike time and the other one for the gids.

[alex4200]

@Hjorthmedh : Any news on this issue?

[Hjorthmedh]

Have worked on converting synapse coordinates to section ID and section X used by Neurodamus. Most recently wrote a plot tool to visually inspect the conversion.

This week and probably a while longer will be working on synaptic input, more specifically how to best create it and determine input locations.

[Hjorthmedh]

Code for generating input written. Next step exporting it.

[jdcourcol]

File format GitHub is here: https://github.com/AllenInstitute/sonata

[Hjorthmedh]

Would it be ok to use a csv file for input spikes. What is the neurodamus internal file format? Is it: neuroninputlocA, time1, time2, ... neuroninputlocB, time1, time2, ... Etc

[Hjorthmedh]

Emailed King & friends asking about Neurodamus file input format.

[Hjorthmedh]

Sent one more reminder asking about file format for input spikes.

[alex4200]

@Hjorthmedh Just to clarify: This ticket is about the SONATA file format which you are trying to use in ticket #157 ?

[jdcourcol]

@Hjorthmedh I think this should go to the sonata issues. https://github.com/AllenInstitute/sonata/issues

[jdcourcol]

@Hjorthmedh the spike file format is documented here: https://github.com/AllenInstitute/sonata/blob/master/docs/SONATA_DEVELOPER_GUIDE.md#spike-file

[Hjorthmedh]

The issue is that I want to verify and run the SONATA simulation through Neurodamus. I have already a first version of the SONATA files created.

That Allen Brain Institute tracker you linked to is more about the SONATA format, and not Neurodamus' ability to read the format?