Spiking cell activity plots

[carolinafernandezp]

Hello,

I would like to plot the spiking cell activity from the spk.txt output files.

I have two questions:

1. For a total of 10 trials, HNN's output looks like this: spk.txt, spk_0.txt ... spk_9.txt
In this case, what is the "spk.txt" file? Is it some sort of average over the 10 trials?

2. The "spk" files consist of two columns: time and cell identifiers. Could you tell me the ranges of the cell identifiers for each type of cell (i.e., layer 2/3 pyramidal, layer 2/3 basket, layer 5 pyramidal, layer 5 basket)? That way I can color them appropriately in my plot.

Thanks in advance,
Carolina

[rythorpe]

1. For a total of 10 trials, HNN's output looks like this: spk.txt, spk_0.txt ... spk_9.txt In this case, what is the "spk.txt" file? Is it some sort of average over the 10 trials?

Yes, I believe the "spk.txt" file is some sort of trial average. @ntolley correct me if I'm wrong here....

2. The "spk" files consist of two columns: time and cell identifiers. Could you tell me the ranges of the cell identifiers for each type of cell (i.e., layer 2/3 pyramidal, layer 2/3 basket, layer 5 pyramidal, layer 5 basket)? That way I can color them appropriately in my plot.

Check out this example. I think it will clarify things for you and provide a straightforward way to plot the spiking activity.

[ntolley]

@carolinafernandezp I can't reproduce this on my end. Do you mind sending a minimal example script?

The expected result is that passing the string fname='spike_%d' will print out a separate file for each trial. Is it possible you ran the code cell multiple times with different strings?

I'm noticing that when the %d is not used, the files are actually overwritten. @rythorpe @jasmainak perhaps this falls under "good first issue" territory to improve the documentation, and raise a warning if the wildcard %d is missing when n_trials > 1

Lastly the attribute net.gid_ranges will give you the cell identifiers.