Convert h5ad file to seurat file

[sanderslhc]

Hi, I 'm using sceasy to convert my h5ad file to seurat file, but there are some problem. When I run the code "sceasy::convertFormat("Blood.h5ad", from="anndata", to="seurat",outFile='Blood.rds')", it will be stopped with "Error: ModuleNotFoundError: No module named 'pandas.core.index'". I have already update my panda to the newest but it still doesn't work. How can I fix it?

[prete]

I think it's the otherway around you need to downgrande your pandas, try pip install -U pandas==1.5.3

[jestlin15]

Hi I faced the same issue and did pip install -U pandas==1.5.3 . however I got ImportError: cannot import name 'Float64Index' from 'pandas.core.indexes.api' after trying to run the convertFormat code again. May I know how I can fix this please? Thank you.

[hopehealey]

I received the same error after switching to pandas 1.5.3. Is there a different pandas version we should use?

[kvegesan-stjude]

Hello, I'm just commenting for future users. pandas==1.3.5 worked for me. After version 1.4, Float64Index was deprecated.

[zhijunyuu]

Hi,

I tried pandas==1.5.3 and pandas==1.3.5, both gave me the same error: cannot import name 'Float64Index' from 'pandas.core.indexes.api'
any idea what I should do to fix it?
Thanks

[sinanugur]

Hi,

Try pandas==1.5.4 anndata==0.9.1

Cheers

[kvegesan-stjude]

This is my environment

anndata==0.6.22.post1
Bottleneck==1.3.5
Brotli==1.0.9
certifi==2023.11.17
cffi==1.16.0
charset-normalizer==2.0.4
cryptography==41.0.7
h5py==3.9.0
idna==3.4
loompy==2.0.16
mkl-fft==1.3.8
mkl-random==1.2.4
mkl-service==2.4.0
natsort==7.1.1
numexpr==2.8.4
numpy==1.24.3
packaging==23.1
pandas==1.3.5
pip==23.3.1
platformdirs==3.10.0
pooch==1.7.0
pycparser==2.21
pyOpenSSL==23.2.0
PySocks==1.7.1
python-dateutil==2.8.2
pytz==2023.3.post1
requests==2.31.0
scipy==1.10.1
setuptools==68.2.2
six==1.16.0
tzdata==2023.3
urllib3==1.26.18
wheel==0.41.2
win-inet-pton==1.1.0

[sinanugur]

This is my environment

anndata==0.6.22.post1 Bottleneck==1.3.5 Brotli==1.0.9 certifi==2023.11.17 cffi==1.16.0 charset-normalizer==2.0.4 cryptography==41.0.7 h5py==3.9.0 idna==3.4 loompy==2.0.16 mkl-fft==1.3.8 mkl-random==1.2.4 mkl-service==2.4.0 natsort==7.1.1 numexpr==2.8.4 numpy==1.24.3 packaging==23.1 pandas==1.3.5 pip==23.3.1 platformdirs==3.10.0 pooch==1.7.0 pycparser==2.21 pyOpenSSL==23.2.0 PySocks==1.7.1 python-dateutil==2.8.2 pytz==2023.3.post1 requests==2.31.0 scipy==1.10.1 setuptools==68.2.2 six==1.16.0 tzdata==2023.3 urllib3==1.26.18 wheel==0.41.2 win-inet-pton==1.1.0

Yeah, anndata==0.9.1 is working, we just tested. and 0.6 returns an error. I recommend

pip install pandas==1.5.4 anndata==0.9.1

[Lil-Psilocybe]

pip install pandas==1.5.4 anndata==0.9.1

When I tried I got ERROR: No matching distribution found for pandas==1.5.4

After looking online it says I may need to change my version of python, but how do I know which version of python works with this anndata and pandas? Thanks, I will try 1.5.3 in the meantime

[prete]

For h5ad to Seurat you should try https://github.com/cellgeni/schard
It's native R so you don't need additionaly python libraries.

[Lil-Psilocybe]

Thanks for a quick reply! I actually got the ModuleNotFoundError: No module named 'pandas.core.index'" when I was converting seurat rds to scanpy h5ad, but there were no issue queries with this error during that type of conversion. I got the h5ad to generate when running pip install pandas==1.5.3 anndata==0.9.1