search

wheaton5 / souporcell

Clustering scRNAseq by genotypes
MIT License
165 stars 46 forks source link

consensus.py Run Error #99

Open zengh58 opened 3 years ago

zengh58 commented 3 years ago

Hello!

When I ran the consensus.py, I met an error : ' Traceback (most recent call last): File "./consensus.py", line 166, in sm = pickle.load(open(os.path.realpath(file)[0:-12]+"stan_consensus.pickle",'rb')) _pickle.UnpicklingError: pickle data was truncated ' Could it be because the messy code in first few rows of the file 'stan_consensus.pickle' make it act as a binary file?

wheaton5 commented 3 years ago

Very unlikely. This package is well used and nobody else has had this issue. Sounds like a corrupted file on clone/download. If not that then maybe some issue i made with the singularity build file I made??

wheaton5 commented 3 years ago

are u using singularity or custom install? custom install requires a specific version of pystan.

zengh58 commented 3 years ago

I met some error with singularity. So I turn to take the hard install method, and run things step by step not through the pipeline script. Would you mind telling me the right version of pystan? I'd like to try to download it. Thank you!

wheaton5 commented 3 years ago

I recommend using the conda environment which should give you the right version of every package.

conda env create -f /path/to/souporcell/souporcell_env.yaml conda activate souporcell

zengh58 commented 3 years ago

Thank you for your advice. But I did exactly use the conda environment 'souporcell'. And I did run it step by step as your pipeline and the code was doing well until the last step that I ran the consensus.py.