cummerbund tool error

Question: cummerbund tool error

7 months ago by

eculfa • 10

Turkey/Istanbul/Üsküdar American Academy

eculfa • 10 wrote:

Hello,

I recently tried differential gene expression analysis using an established workflow. I had miRNA data of two conditions, single replicate. I fed my fastqsanger data and the reference genome I called from UCSC, into the workflow. Galaxy did generate a Cuffdiff SQlite database but when I use it to get plots with cummerbund tool, I receive the error below:

cummerbund tool error "Attempting to run a tool with empty command definition."

The workflow I used is; DEG:Tuxedo2 - 2 Conditions 1 Replicate each

What do you think is the problem? Thank you!! Efraim

EDIT: I just checked the other data generated by the workflow and I have two other errors from Cufflinks and Cuffmerge; "An error occurred setting the metadata for this dataset Set it manually or retry auto-detection" Could they be connected?

rna-seq cufflinks galaxy workflows • 350 views

ADD COMMENT • link •

modified 7 months ago by Jennifer Hillman Jackson ♦ 25k • written 7 months ago by eculfa • 10

7 months ago by

Jennifer Hillman Jackson ♦ 25k

United States

Jennifer Hillman Jackson ♦ 25k wrote:

Hello,

Yes, the problems are likely related. Double check your inputs. If all check out OK, then you could consider editing the workflow to use the most current version of the included tools. CummeRbund had some problems earlier but the tool has been updated, as have most of the others in this suite. Please note that Tophat was not updated, is deprecated, should be avoided, with HISAT2 used instead as the replacement tool.

The Galaxy support FAQs cover many common format/content issues that can lead to tool errors: https://galaxyproject.org/support/#troubleshooting

Start here: My job ended with an error. What can I do? https://galaxyproject.org/support/tool-error/
The above FAQ includes help links to sub-topics with details, most of which are listed in this section: https://galaxyproject.org/support/#getting-inputs-right

I replied to your bug report about this same problem with a link back here. We can't help to troubleshoot the inputs in details, as they were permanently deleted, but I can see a few problems. A Cuffcompare GTF was input to Cuffdiff instead of the GTF from Cuffmerge -- and this won't have quite the same content. Most issues with using this tool suite are due to custom genome formatting problems, mismatches between the reference annotation/genome, reference GTF annotation sourced from UCSC (use iGenomes instead, why with details https://biostar.usegalaxy.org/p/21827/#21845 -- pick the UCSC/hg38 archive), the wrong database accidentally mapped against, and insufficient fastq QA/QC leading to poor mapping results.

All of that said, current RNA-seq DE analysis tools/methods/workflows are covered in the Galaxy Tutorials. https://galaxyproject.org/learn/

Start here if interested: RNA-seq: Discovering and quantifying new transcripts - an in-depth transcriptome analysis example https://galaxyproject.org/tutorials/nt_rnaseq/
HISAT2 can be used to map data for use with the Cuff* tools if you only choose to only swap out that tool for the deprecated Tophat tool. Instructions are here in the above tutorial. Select output for Cufflinks, instead of Stringtie, same tool form section: https://galaxyproject.org/tutorials/nt_rnaseq/#spliced-mapping-with-hisat

Thanks! Jen, Galaxy team

ADD COMMENT • link modified 7 months ago • written 7 months ago by Jennifer Hillman Jackson ♦ 25k

Similar posts • Search »