multiple Visium sections in one adata object does not work with `sc.pl.spatial(img_key=None)`

Does scanpy==1.5.1 support multiple sections in one adata object? If I concatenate several anndata object I can’t plot even with sc.pl.spatial(img_key=None).

Try concatenating 3 mouse brain adata object and plotting:

adata = adata1.concatenate([obj2, obj3], index_unique=None)
sc.pl.spatial(adata[adata.obs["sample"]==adata.obs["sample"].unique()[0], :], 
              color=["Rorb", "Vip"], img_key=None,
              vmin=0, cmap='magma',
              gene_symbols='SYMBOL')

This is the error I get:

---------------------------------------------------------------------------
KeyError                                  Traceback (most recent call last)
<ipython-input-9-8c84185773ec> in <module>
      6               color=["Rorb", "Vip"], img_key=None,
      7               vmin=0, cmap='magma', #vmax=3.8,
----> 8               gene_symbols='SYMBOL'
      9              )

/nfs/team283/vk7/software/miniconda3farm5/envs/cellpymc/lib/python3.7/site-packages/scanpy/plotting/_tools/scatterplots.py in spatial(adata, img_key, library_id, crop_coord, alpha_img, bw, size, **kwargs)
    765     """
    766     if library_id is _empty:
--> 767         library_id = next((i for i in adata.uns['spatial'].keys()))
    768     else:
    769         if library_id not in adata.uns['spatial'].keys():

KeyError: 'spatial'

Versions:

scanpy==1.5.1 anndata==0.7.1 umap==0.3.10 numpy==1.17.3 scipy==1.4.1 pandas==0.25.3 scikit-learn==0.22.1 statsmodels==0.10.2 python-igraph==0.7.1 louvain==0.6.1 leidenalg==0.7.0

[ivirshup]

Much of the spatial data is stored in uns, which does not get combined by default.

There is an example of concatenating visium datasets in the tutorial and more information on concatenating .uns in the latest anndata docs.

[ivirshup]

No problem!