Why is the group apic used in the fitting data but not present in the cell?

[pgleeson]

The chosen all active cell has fitted parameters which use group apic: https://github.com/OpenSourceBrain/AllenInstituteNeuroML/blob/af1a0baeeee029a09bca7d5f9cd809eda55adaf8/CellTypesDatabase/models/497233223/464198958_fit.json#L249

These are usually defined by type (2nd column in SWC file) = 4, e.g. https://github.com/OpenSourceBrain/AllenInstituteNeuroML/blob/af1a0baeeee029a09bca7d5f9cd809eda55adaf8/CellTypesDatabase/models/483108201/Rbp4-Cre_KL100_Ai14-196623.04.01.01_491119548_m.swc#L1185

but this is not present in the swc file for this cell: https://github.com/OpenSourceBrain/AllenInstituteNeuroML/blob/pg_test/CellTypesDatabase/models/497233223/Sst-IRES-Cre_Ai14-165865.03.01.01_491119181_m.swc

The 3d view also suggests no apic: http://celltypes.brain-map.org/experiment/morphology/464198958

So why is it used in the fit file??

[pgleeson]

Quote from http://help.brain-map.org/download/attachments/8323525/BiophysModelAllActive.pdf backs this up: "The compartments of excitatory neurons were separated into axon initial segment (AIS), soma, basal dendrites, and apical dendrites. For inhibitory neurons, there was only a single dendritic zone."

[pgleeson]

@kaeldai This is probably not your area, but just wondering if you know of someone we could ask... Basically, why would an interneuron cell model with no apical dendrite have values in its tuned parameter file specifically for the apical dendrite group?

[kaeldai]

Hi @pgleeson - I don't know much about the all-active models or how they were optimized. I think the best person to ask would be Nathan Gouwens who did a lot of the main cell-types db optimization. I don't know if he's on github but you can try him at nathang at alleninstitute dot org.

[pgleeson]

Thanks @kaeldai! Will do.