Closed pakiessling closed 1 year ago
Hi @pakiessling , we are working on a workflow with the recently announced SpatialData data structure, which we think is an effective way to store large spatial omics datasets with support for storing molecule and cell segmentation data we use in Bento. I would look at SpatialData-IO for ingesting datasets. Stay tuned for updates 😄
If you can't wait and want to get your hands dirty, we started writing some functions that deal with parsing parquet
files from Xenium data for shapes but they are not available yet. You can find them here, but it is messy. Hope that helps for now.
https://github.com/ckmah/bento-tools/blob/0021f59b9c644cfdf9bba1143e06f8bdfadc3541/bento/io/_io.py#L470
@ckmah Perfect, thanks a lot
Thank you for the tool, exited to try it out.
Could you give a humble biologist some pointers on how to load in Merfish output data? The segmentation is saved as geopandas in a .parquet file.
I assume I need to use
bento.io.prepare
andrasterio
for this, but it is not entirely clear to me.