Question: DeepTools plotCorrelation Error "IOError: [Errno 22] Invalid argument"
1
gravatar for alim
4 months ago by
alim10
United Kingdom
alim10 wrote:

Dear all,

I used a matrix generated from multiBamSummary to run plotCorrelation but the error message below was generated:

Fatal error: Exit code 1 () Traceback (most recent call last): File "/galaxy-repl/main/jobdir/014/089/14089566/conda-env/bin/plotCorrelation", line 7, in <module> main() File "/galaxy-repl/main/jobdir/014/089/14089566/conda-env/lib/python2.7/site-packages/deeptools/plotCorrelation.py", line 172, in main skip_zeros=args.skipZeros) File "/galaxy-repl/main/jobdir/014/089/14089566/conda-env/lib/python2.7/site-packages/deeptools/correlation.py", line 32, in __init__ self.load_matrix(matrix_file) File "/galaxy-repl/main/jobdir/014/089/14089566/conda-env/lib/python2.7/site-packages/deeptools/correlation.py", line 72, in load_matrix _ma = np.load(matrix_file) File "/galaxy-repl/main/jobdir/014/089/14089566/conda-env/lib/python2.7/site-packages/numpy/lib/npyio.py", line 392, in load fid.seek(-N, 1) # back-up IOError: [Errno 22] Invalid argument

Could anyone help please?

Many thanks, Alfred

ADD COMMENTlink written 4 months ago by alim10

That's a new one. What's the size of the item in your history? My only guess is that it's ~0 bytes.

ADD REPLYlink written 4 months ago by Devon Ryan1.2k

Hi Devon, it is 0 bytes indeed. Any advice on how to get over this would be much appreciated !

ADD REPLYlink written 4 months ago by alim10

That suggests that something went wrong with multiBamSummary. I would normally expect that to have produced an error message (perhaps on "stderr" if you click on the info button on the history item). Can you (A) see if there's any apparent error message that multiBamSummary produced and (B) post the settings you used?

ADD REPLYlink written 4 months ago by Devon Ryan1.2k

Hi Devon, I just double checked and you're right, the multiBamSummary file was actually an empty file with 0 byte of data ! I should have checked before assuming it was okay and put it into plotCorrelation. There is however no error message from the multiBamSummary.

When I ran mBS, my input files were TopHat Bam files, generated using alignment to the mm10 reference genome. I have four samples that I wanted to compare, so my mBS were generated using all 4 TopHat files as inputs. I used all the mBS default settings as below: 1. Computational mode = Bin 2. Bin size in bp = 10,000 3. Distance between bins = 0 4. Region of the genome to limit the operation to = Left blank 5. Show advanced options = No 6. Save raw counts (coverages) to file = No

Many thanks !

ADD REPLYlink written 4 months ago by alim10

Hmm, I don't have any good guesses then. What version of deepTools are you using in Galaxy? Can you somehow provide me with the BAM files? I can then track this down. I'm planning the next deepTools release in a couple weeks, so if this is a bug I'd like to get it resolved beforehand :)

ADD REPLYlink written 4 months ago by Devon Ryan1.2k

Hi Devon, it is Galaxy Version 2.3.5.0. What would be the best way to send you those BAM files? Many thanks for looking into it. I would really like to analyse these data before the end of the week if possible

ADD REPLYlink written 4 months ago by alim10

Depending on their size, you might be able to:

  1. Create an account on our public deepTools Galaxy server
  2. Upload the BAM files there via FTP.
  3. Either email me or just post here alluding to the email address you used to register, so I can find the files.

The fewer files the better, just for the sake of speed, so if you have the same problem just using 2 BAM files then just upload those two.

ADD REPLYlink written 4 months ago by Devon Ryan1.2k

We are getting the same error on the public Galaxy server http://usegalaxy.org. Same DeepTools version. Am looking at it to see if there is a problem with inputs now. Also interested in what happens here - if true tool issue, DeepTools is updated, etc. Thanks! Jen, Galaxy team

ADD REPLYlink written 4 months ago by Jennifer Hillman Jackson21k

It's weird, the history was shared with me and when I run this from the command line with the files used I get output. I'll upload the files to our Galaxy server and see if I can reproduce the issue.

ADD REPLYlink written 4 months ago by Devon Ryan1.2k

Cool - thanks. Can share history with me directly if can reproduce error - use the bx address. Tx!

ADD REPLYlink written 4 months ago by Jennifer Hillman Jackson21k

Will do.

ADD REPLYlink written 4 months ago by Devon Ryan1.2k

It should be shared now.

ADD REPLYlink written 4 months ago by Devon Ryan1.2k

OK, I can reproduce this on our Galaxy instance too, which suggests that it's a wrapper bug. I'll try to figure that out and have Björn push out the updated wrapper tomorrow.

ADD REPLYlink written 4 months ago by Devon Ryan1.2k
1
gravatar for Devon Ryan
4 months ago by
Devon Ryan1.2k
Germany
Devon Ryan1.2k wrote:

@alim: This is now bug #436 github. I think I've found the issue, which I'm surprised no one has reported before.

ADD COMMENTlink written 4 months ago by Devon Ryan1.2k

ah well done Devon ! Many thanks ! Would it be possible for me to try re-running multiBamSummary on Galaxy again anytime soon?

ADD REPLYlink written 4 months ago by alim10

Watch the ticket for the core correction.

As the correction is published, installed, and tested at http://usegalaxy.org, problems like this move between categories on this supplemental support wiki: https://github.com/jennaj/support-known-issues/wiki/Tool-Issues

ADD REPLYlink written 4 months ago by Jennifer Hillman Jackson21k
Please log in to add an answer.

Help
Access

Use of this site constitutes acceptance of our User Agreement and Privacy Policy.
Powered by Biostar version 16.09
Traffic: 74 users visited in the last hour