ventolab/CellphoneDB

Unable to run cellphonedb due to database

molybdenumt opened this issue · 7 comments

molybdenumt commented

I'm using R with the conda environment

module load anaconda3
module load R
conda create -n cpdb python=3.7
source activate cpdb

cellphonedb database download

I kept getting:
There are no versions available (or connection could not be made to server to retrieve them)
So I downloaded the cellphonedb.zip from https://github.com/ventolab/CellphoneDB-data and move it to the folder /u/.cpdb/releases/v5.0.0
Then I ran this code
cellphonedb method degs_analysis cellphonedb_mergeb_meta.txt cellphonedb_mergeb_scaledata.txt mergeb_markerct.txt --verbose
But keep getting this:

[ ][APP][05/11/23-20:26:00][WARNING] Database '{value}' is not available. Trying to download it.
There are no versions available (or connection could not be made to server to retrieve them)

I also tried this
cellphonedb method statistical_analysis cellphonedb_mergeb_meta.txt cellphonedb_mergeb_scaledata.txt --verbose
And got the same error:

[ ][APP][05/11/23-20:35:18][WARNING] Database '{value}' is not available. Trying to download it.
There are no versions available (or connection could not be made to server to retrieve them)
prete commented

Hi @molybdenumt thank you for using CellPhoneDB, could I check which version of CPDB is installed in your environment?

Python 3.7 will pull CellPhoneDB ~3. Any version of CellPhoneDB after 3.0.1 won't be compatible with rercent versions of python nor the databse. You should create your environment with

conda create -n cpdb python=3.9

For recent versions of the tool has also moved away from the command line towards a python/notebook use of the library. The most practical thing for you should be create a python file like this one:

from cellphonedb.src.core.methods import cpdb_statistical_analysis_method

cpdb_results = cpdb_statistical_analysis_method.call(
        cpdb_file_path = "/u/.cpdb/releases/v5.0.0/cellphonedb.zip",
        meta_file_path = "cellphonedb_mergeb_meta.txt",
        counts_file_path = "cellphonedb_mergeb_scaledata.txt",
        counts_data = "hgnc_symbol",
        score_interactions = True,
        threshold = 0.1,
        output_path = "outs")

name it cpdb_run.py and then execute it using python cpdb_run.py

molybdenumt commented

Thank you for your response. When I tried running the code, it has a new error:

Traceback (most recent call last): File "/u/home/t/trando/cpdb_run.py", line 4, in <module> cpdb_results = cpdb_statistical_analysis_method.call( TypeError: call() got an unexpected keyword argument 'score_interactions'

datasome commented

Hi molybdenumt,

If I may just chime in here - once you created your clean environment with:
conda create -n cpdb python=3.9
did you install CellphoneDB via:
pip install cellphonedb ?
From the above error it looks as if you've added score_interactions argument to the call in cpdb_statistical_analysis_method.call() in cpdb_run.py file, but the version of CellphoneDB you installed in your environment is not the latest one. Could you please confirm?

Best,

Robert.

molybdenumt commented
datasome commented

Thanks molybdenumt,

I cannot quite see how this can happen..
Could I please ask you to send me the precise list of commands you used to first remove the previous conda cpdb environment, then re-creating it, installing cellphonedb then running your script; and also the exact content of your cpdb_run.py script?
Alternatively you can send me a Zoom link on contact@cellphonedb.org and you can show me what you're doing step by step.

Best wishes,

Robert.

molybdenumt commented

Hi Robert,

I sent the email, but here is more updates:
conda deactivate conda remove -n cpdb –all conda create -n cpdb python=3.9 conda activate cpdb pip install cellphonedb vi cpdb_run.py 

from cellphonedb.src.core.methods import cpdb_degs_analysis_method
cpdb_results = cpdb_degs_analysis_method.call(
        cpdb_file_path = "/u/home/t/.cpdb/releases/v5.0.0/cellphonedb.zip",
        meta_file_path = "cellphonedb_mergeb_meta.txt",
        counts_file_path = "cellphonedb_mergeb_scaledata.txt",
        counts_data = "hgnc_symbol",
        score_interactions = True,
        threshold = 0.1,
        output_path = "/outs")

Still getting error :

Traceback (most recent call last):
  File "/u/home/t/cpdb_run.py", line 4, in <module>
    cpdb_results = cpdb_statistical_analysis_method.call(
TypeError: call() got an unexpected keyword argument 'score_interactions'

When I removed the 2 lines since this was what was written on this website https://cellphonedb.readthedocs.io/en/latest/RESULTS-DOCUMENTATION.html#method-2-statistical-inference-of-interaction-specificity

        score_interactions = True,
        threshold = 0.1,

Now it's only

from cellphonedb.src.core.methods import cpdb_degs_analysis_method
cpdb_results = cpdb_degs_analysis_method.call(
        cpdb_file_path = "/u/home/t/.cpdb/releases/v5.0.0/cellphonedb.zip",
        meta_file_path = "cellphonedb_mergeb_meta.txt",
        counts_file_path = "cellphonedb_mergeb_scaledata.txt",
        counts_data = "hgnc_symbol",
        output_path = "/outs")

It ran and now got a new error:

Reading user files...
Traceback (most recent call last):
  File "/u/home/t/cpdb_run.py", line 4, in <module>
    cpdb_results = cpdb_statistical_analysis_method.call(
  File "/u/home/t/.local/lib/python3.9/site-packages/cellphonedb/src/core/methods/cpdb_statistical_analysis_method.py", line 98, in call
    counts, meta, microenvs, degs = file_utils.get_user_files( \
  File "/u/home/t/.local/lib/python3.9/site-packages/cellphonedb/utils/file_utils.py", line 315, in get_user_files
    counts = read_data_table_from_file(counts_fp, index_column_first=True)
  File "/u/home/t/.local/lib/python3.9/site-packages/cellphonedb/utils/file_utils.py", line 48, in read_data_table_from_file
    return _read_data(f, separator, index_column_first, dtype, na_values, compression)
  File "/u/home/t/.local/lib/python3.9/site-packages/cellphonedb/utils/file_utils.py", line 106, in _read_data
    return pd.read_csv(file_stream, sep=separator, index_col=0 if index_column_first else None, dtype=dtype,
  File "/u/home/t/.local/lib/python3.9/site-packages/pandas/io/parsers/readers.py", line 912, in read_csv
    return _read(filepath_or_buffer, kwds)
  File "/u/home/t/.local/lib/python3.9/site-packages/pandas/io/parsers/readers.py", line 577, in _read
    parser = TextFileReader(filepath_or_buffer, **kwds)
  File "/u/home/t/.local/lib/python3.9/site-packages/pandas/io/parsers/readers.py", line 1407, in __init__
    self._engine = self._make_engine(f, self.engine)
  File "/u/home/t/.local/lib/python3.9/site-packages/pandas/io/parsers/readers.py", line 1679, in _make_engine
    return mapping[engine](f, **self.options)
  File "/u/home/t/.local/lib/python3.9/site-packages/pandas/io/parsers/c_parser_wrapper.py", line 93, in __init__
    self._reader = parsers.TextReader(src, **kwds)
  File "pandas/_libs/parsers.pyx", line 555, in pandas._libs.parsers.TextReader.__cinit__
pandas.errors.EmptyDataError: No columns to parse from file
molybdenumt commented

For anyone having the same issues, I removed the environment and reinstalled the package with the team's help.

conda env remove -n cpdb_docs
conda create -n cpdb python=3.9
conda activate cpdb
pip install --force-reinstall cellphonedb

create a file called cpdb.py (as below) and run the file

from cellphonedb.src.core.methods import cpdb_degs_analysis_method
 

if __name__ == '__main__':

        cpdb_results = cpdb_degs_analysis_method.call(

                       cpdb_file_path = "~/.cpdb/releases/v5.0.0/cellphonedb.zip",

                       meta_file_path = "./example_data/test_meta.txt",

                       counts_file_path = "./example_data/test_counts.txt",

                       degs_file_path = "./example_data/test_degs.txt",

                       counts_data = "ensembl",

                       score_interactions = True,

                       threshold = 0.1,

                       output_path = "./outs")