DiffBind ln failed to create symbolic link file-bamreads.bam File exists

Question: DiffBind ln failed to create symbolic link file-bamreads.bam File exists

24 days ago by

Psyduck • 0 wrote:

Hi there,

I am trying to use DiffBind (Galaxy Version 2.6.6.4) to combine ChIP-seq replicates and compare between samples. Although the job runs, the output has 0 lines and below, the error: ln: failed to create symbolic link ‘bio1-bamreads.bam’: File exists

I am using the BAM files (simple names without any characters or spaces) and MACS2 peak files that I get directly from BWA-MEM mapping and MACS2 as outputs.

I'm almost sure that this is something small like changing the file format or name but I have tried everything I can think of so can anyone help please?

symbolic link diffbind bam • 105 views

ADD COMMENT • link •

modified 23 days ago • written 24 days ago by Psyduck • 0

23 days ago by

Jennifer Hillman Jackson ♦ 25k

United States

Jennifer Hillman Jackson ♦ 25k wrote:

Hi,

This sounds like a cluster error. Try rerunning if you have not already.

If that fails the same way again, please report the error using the bug icon if working at Galaxy Main https://usegalaxy.org or can reproduce the problem there. Please include a link to this post in the comments so we can associate the two. For other servers, you'll need to contact the administrators that run it and report the issue. Contact information is usually on the home page of the server.

Thanks! Jen, Galaxy team

ADD COMMENT • link written 23 days ago by Jennifer Hillman Jackson ♦ 25k

23 days ago by

Psyduck • 0

Psyduck • 0 wrote:

Hi Jen,

Thank you for your reply.

I tried running it again today and got the same error. I am not able to submit a bug report. Although the output contains an error, the job still appears green as if successful and therefore no bug icon.

I have submitted the history link to galaxy-bugs@lists.galaxyproject.org

Thanks

ADD COMMENT • link written 23 days ago by Psyduck • 0

Ok, thanks, I'll take a look and get back soon. Jen

ADD REPLY • link written 22 days ago by Jennifer Hillman Jackson ♦ 25k

The issue is with the peak files. These are not really in BED format (although that datatype is assigned). Notice how the counts for comment/data lines do not add up to the actual content (no comment lines, all are actually data lines that are poorly formatted for BED specification). These are also not the result from calling peaks on the BAM files submitted to the tool Deseq2.

How to fix: Call peaks on the BAM files submitted to Deseq2 then use that output for the peak (BED) inputs.

ADD REPLY • link modified 22 days ago • written 22 days ago by Jennifer Hillman Jackson ♦ 25k

Please log in to add an answer.

Similar posts • Search »