heilandd/NFCN2

unable to import s3 and s5 .h5ad objects [GBM spatial paper; Nature Communications 2022]

Opened this issue · 3 comments

dr-sayyadhury commented

Hi, I was interested to look further into the Spatial Transcriptomics datasets from 3 patient-derived GBM samples that was recently published last week in Nature Communication. However, I am running into problems importing 2 out of your 3 datasets. Below are the error codes I receive.

Sample S3
---------s3=sc.read_h5ad('/Users/ShaminiAyyadhury/Research/Projects_Experiments/SmartTherapeuticsGBM/ST_spatial/osfstorage-archive/S3/S3.h5ad')
Traceback (most recent call last):

File "/var/folders/xl/40x0m_b12y5fz7w2hqr_yf480000gp/T/ipykernel_12766/4188748622.py", line 1, in
s3=sc.read_h5ad('/Users/ShaminiAyyadhury/Research/Projects_Experiments/SmartTherapeuticsGBM/ST_spatial/osfstorage-archive/S3/S3.h5ad')

File "/Users/ShaminiAyyadhury/anaconda3/envs/spatial/lib/python3.9/site-packages/anndata/_io/h5ad.py", line 437, in read_h5ad
return AnnData(**d)

File "/Users/ShaminiAyyadhury/anaconda3/envs/spatial/lib/python3.9/site-packages/anndata/_core/anndata.py", line 308, in init
self._init_as_actual(

File "/Users/ShaminiAyyadhury/anaconda3/envs/spatial/lib/python3.9/site-packages/anndata/_core/anndata.py", line 526, in _init_as_actual
self._check_dimensions()

File "/Users/ShaminiAyyadhury/anaconda3/envs/spatial/lib/python3.9/site-packages/anndata/_core/anndata.py", line 1842, in _check_dimensions
raise ValueError(

ValueError: Variables annot. var must have number of columns of X (21743), but has 21643 rows.

Sample S5

s5=sc.read_h5ad('/Users/ShaminiAyyadhury/Research/Projects_Experiments/SmartTherapeuticsGBM/ST_spatial/osfstorage-archive/S5/S5.h5ad')
Traceback (most recent call last):

File "/var/folders/xl/40x0m_b12y5fz7w2hqr_yf480000gp/T/ipykernel_12766/137605449.py", line 1, in
s5=sc.read_h5ad('/Users/ShaminiAyyadhury/Research/Projects_Experiments/SmartTherapeuticsGBM/ST_spatial/osfstorage-archive/S5/S5.h5ad')

File "/Users/ShaminiAyyadhury/anaconda3/envs/spatial/lib/python3.9/site-packages/anndata/_io/h5ad.py", line 437, in read_h5ad
return AnnData(**d)

File "/Users/ShaminiAyyadhury/anaconda3/envs/spatial/lib/python3.9/site-packages/anndata/_core/anndata.py", line 308, in init
self._init_as_actual(

File "/Users/ShaminiAyyadhury/anaconda3/envs/spatial/lib/python3.9/site-packages/anndata/_core/anndata.py", line 526, in _init_as_actual
self._check_dimensions()

File "/Users/ShaminiAyyadhury/anaconda3/envs/spatial/lib/python3.9/site-packages/anndata/_core/anndata.py", line 1842, in _check_dimensions
raise ValueError(

ValueError: Variables annot. var must have number of columns of X (21743), but has 21643 rows.

Would be grateful if you can look into this and let me know what the issue might be.

Thank you for your time.

Regards,
Shamini A

heilandd commented
dr-sayyadhury commented

Sure ... thank you.

dr-sayyadhury commented

Hi, any luck with trying to sort out this issue? I tried to independently convert the Seurat objects to AnnData objects. But it gave the same error as well. I really need to run my scripts on AnnData format as I am using fabian theis's Squidpy package, which runs on AnnData. Thank you.