Concating dfs pandas error

[saycheeeeese]

Hi I have received the following error

Concating dfs. (File: run_epic, Log level: INFO, Time: Wed, 24 May 2017 14:50:32 )
Traceback (most recent call last):
  File "/home/ubuntu/.local/bin/epic", line 219, in <module>
    run_epic(args)
  File "/usr/local/lib/python2.7/dist-packages/bioepic-0.1.25-py2.7.egg/epic/run/run_epic.py", line 60, in run_epic
    df = pd.concat([df for df in dfs if not df.empty])
  File "/usr/local/lib/python2.7/dist-packages/pandas/tools/merge.py", line 845, in concat
    copy=copy)
  File "/usr/local/lib/python2.7/dist-packages/pandas/tools/merge.py", line 878, in __init__
    raise ValueError('No objects to concatenate')
ValueError: No objects to concatenate

I am unsure if this an error in pandas or epic? Any ideas of how to fix this? Thanks

[endrebak]

Could you print the whole epic log and the error message?

I think the problem ultimately comes from the fact that your custom chromosome file and bed files contain different chromosome names, so if you could send the output of

head <your_bed_files> and cat <your_chromsizes_file> it would be appreciated.

So if your chromsizes file is called build/lib/epic/scripts/chromsizes/dm3.chromsizes you call cat build/lib/epic/scripts/chromsizes/dm3.chromsizes and print the output here. Same with the bed files you use please, only make sure to use head instead of cat.

If this isn't it, I'd appreciate if you could make the files available on dropbox or something so I could reproduce the error. You should send the links in an e-mail if you want to keep the files private.

[endrebak]

Epic might also not have found any enriched regions in your data, that is another possibility, but sounds unlikely to me. Could you print the whole command you use to invoke epic please?

Note to self: Add warning message if all dfs are empty before concating. Should go here: https://github.com/biocore-ntnu/epic/blob/master/epic/run/run_epic.py#L57 and matrix creation should still work afterwards.