search

BlueBrain / atlas-densities

Tools to compute densities in the context of brain atlases.
Apache License 2.0
2 stars 7 forks source link

Hard coded E-type in scale_excitatory_densities #57

Closed KeremKurban closed 3 months ago

KeremKurban commented 5 months ago

In splitting exc densities into metypes I see cADpyr is hard coded. There might be other cases, so passing it as an argument can help generalize to noncortical circuits

mgeplf commented 5 months ago

Interesting. The code was provided by @dkeller9 in https://github.com/BlueBrain/atlas-densities/pull/15/, and https://github.com/BlueBrain/atlas-densities/pull/19, I believe.

@dkeller9 are there other changes that are required to make it less region specific?

dkeller9 commented 5 months ago

I agree that passing it as an argument is better if it is intended to be less region-specific. I thought this was a safe choice since the etype of cortical excitatory cells is unlikely to change. I think this should be the only modification needed.

mgeplf commented 5 months ago

Before I make the change, @KeremKurban is it necessary if one is use the create-from-probability-map instead of excitatory-split?

KeremKurban commented 5 months ago

Do you mean is there any noncortical circuit that wants use excitatory-split ?

mgeplf commented 5 months ago

Do you mean is there any noncortical circuit that wants use excitatory-split ?

Yeah, is it possible to use create-from-probability-map; that way all areas use the same code, and there aren't special cases.

mgeplf commented 3 months ago

@KeremKurban has this been resolved?

KeremKurban commented 3 months ago

If i remember correctly, the excitatory cell types will migrate to create-from-probability-map approach at some point. @lecriste or @dkeller9 might know better. If we migrate to that option, then this issue can be closed as it wont be used anymore.

mgeplf commented 3 months ago

Yes, create-from-probability-map should be used. I will close this, if it needs to be reopened, we can do that. Thanks.