Atera Pre-process¶

Make DegaFiles from a 10x Atera output bundle. Atera is Xenium-class, so dega.pre.main detects an Atera bundle (by its experiment.spatial file) and runs the Xenium pipeline with Atera readers; view the result with technology="Xenium".

Requires the pre-processing extra (pip install "celldega[pre]", which needs libvips). Each region is one bundle folder, e.g. data/atera/EA_s3_broad_DX002/, containing experiment.spatial, cells.zarr.zip, transcripts.zarr.zip, cell_feature_matrix.zarr.zip, and morphology_2d/.

A large Atera WTA region (~430k cells, 3.4 billion transcripts) takes about an hour, peaks around 55 GB of memory, and produces ~20 GB of DegaFiles.

In [ ]:

import celldega as dega

sample = "EA_s3_broad_DX002"
path_dega_files = f"data/atera_dega_files/{sample}"

dega.pre.main(
    sample=sample,
    data_root_dir="data/atera",
    tile_size=250,
    image_tile_layer="all",
    path_dega_files=path_dega_files,
)

import celldega as dega sample = "EA_s3_broad_DX002" path_dega_files = f"data/atera_dega_files/{sample}" dega.pre.main( sample=sample, data_root_dir="data/atera", tile_size=250, image_tile_layer="all", path_dega_files=path_dega_files, )

In [ ]:

dega.viz.Landscape(
    technology="Xenium",
    base_url=f"http://localhost:{dega.viz.get_local_server()}/{path_dega_files}",
)

dega.viz.Landscape( technology="Xenium", base_url=f"http://localhost:{dega.viz.get_local_server()}/{path_dega_files}", )