Trouble uploading genome

usegalaxy.org support
1

I keep getting this error when I try to upload my data files. The FTP option doesn’t work for my data sets. I’m not sure why this is happening.

Fatal error: Exit code 1 ()
Traceback (most recent call last):
File “/cvmfs/main.galaxyproject.org/galaxy/tools/data_source/upload.py”, line 326, in
main()
File “/cvmfs/main.galaxyproject.org/galaxy/tools/data_source/upload.py”, line 319, in main
metadata.append(add_file(dataset, registry, output_path))
File “/cvmfs/main.galaxyproject.org/galaxy/tools/data_source/upload.py”, line 128, in add_file
convert_spaces_to_tabs=dataset.space_to_tab,
File “/cvmfs/main.galaxyproject.org/galaxy/lib/galaxy/datatypes/upload_util.py”, line 70, in handle_upload
if check_content and hasattr(datatype, ‘sniff’) and not datatype.sniff(path):
File “/cvmfs/main.galaxyproject.org/galaxy/lib/galaxy/datatypes/sniff.py”, line 632, in auto_sniff
return self.sniff_prefix(file_prefix)
File “/cvmfs/main.galaxyproject.org/galaxy/lib/galaxy/datatypes/sequence.py”, line 355, in sniff_prefix
fh = file_prefix.string_io()
File “/cvmfs/main.galaxyproject.org/galaxy/lib/galaxy/datatypes/sniff.py”, line 589, in string_io
raise Exception(“Attempting to create a StringIO object for binary data.”)
Exception: Attempting to create a StringIO object for binary data.

1 Like
2

Hi - Using this option in the Upload tool with compressed data is likely the problem.

Choices:

  • Load uncompressed data
  • Load compressed data, avoiding this option. Instead, run the tool Convert delimiters to TAB once the data is in the history.

If those don’t work, a few questions:

  • What format is the original data in?
  • Compression type?
  • Are you using “autodetect” for the datatype?
  • If not, maybe try that and see what datatype is assigned versus the expected datatype.

Small update: If you are loading a fasta genome (compressed or not), you probably don’t want to convert spaces to tabs (at any step). Instead, you probably want to use “NormalizeFasta”. More about Preparing and using a Custom Reference Genome or Build