iblatlas.genomics.agea

A package for loading 4345 formatted and registered gene expression volumes

Functions

`load`	Reads in the Allen gene expression experiments binary data.
`load_atlas`	Builds the brain atlas matching the AGEA / MERFISH 200 um grid.

load_atlas(folder_cache=None)[source]

Builds the brain atlas matching the AGEA / MERFISH 200 um grid.

Only fetches image.npy and label.npy (a few MB) rather than the full atlas/agea S3 folder, so this is cheap to call even when the gene expression volumes themselves are not needed – e.g. to index a MERFISH volume (iblatlas.genomics.merfish.load_volume).

Parameters:: folder_cache
Returns:: a brainatlas object with the labels and coordinates matching the gene expression / MERFISH volumes

load(folder_cache=None, expression_size=(4345, 58, 41, 67), label='')[source]

Reads in the Allen gene expression experiments binary data. Generation scripts from the Allen Institute are available in the gene-expression-scrapping folder

Parameters:

filename
folder_cache
label – which volume label to load - ‘’: original volumes with no processing applied - ‘processed’: processed volumes with ML symmetry, PPCA imputation and curtaining

Returns:

a dataframe of experiments (4345, 2), where each record corresponds to a single gene expression a memmap of all experiments brain volumes, size (4345, 58, 41, 67) corresponding to

(nexperiments, ml, dv, ap). The spacing between slices is 200 um: a brainatlas object with the labels and coordinates matching the gene expression volumes