Variable doesn't work in Workflow editor for certain tools?

Question: Variable doesn't work in Workflow editor for certain tools?

22 months ago by

Tel Aviv University

Hello,

I've started using variables in the workflow editor to make tracking the datasets easier, and so far its been working OK. One tool has given me trouble, though - UnifiedGenotyper. The BAM input name seems to be stored in the input_bam variable, but when I set the new dataset's name to #{input_bam} - extra text the variable resolves to an empty string. This tool has another varaible, ref_file (which references the name of the FASTA reference file), and that one works. I tried it a number of time, so relatively sure it's not just a typo on my part.

Thanks for any help!

workflow • 684 views

ADD COMMENT • link •

modified 21 months ago by Jennifer Hillman Jackson ♦ 25k • written 22 months ago by vladimir.gritsenko • 90

21 months ago by

Jennifer Hillman Jackson ♦ 25k

United States

Jennifer Hillman Jackson ♦ 25k wrote:

Hello,

Yes, this is a possible outcome for certain tools. In particular, if using GATK tools at http://usegalaxy.org, the wrappers are deprecated and the underlying GATK tools are older (version 1.4). These will not be undergoing further development/tuning and will be at some point removed from the server entirely.

Using GATK in a local or cloud Galaxy, with the latest wrapper/tool versions, is recommended. Should you run into problems with these, create an issue against the Github development repository for the tool describing the unexpected behavior. A link to that can be found on the tool's home page in the Tool Shed.

Thanks!

ADD COMMENT • link written 21 months ago by Jennifer Hillman Jackson ♦ 25k

Thanks for the reply! Is there a particular reason why you're phasing out these Broad tools? Aren't they considered best-of-breed for SNP calling? Does that mean you're recommending FreeBayes and VarScan for that purpose?

ADD REPLY • link written 21 months ago by vladimir.gritsenko • 90

The licensing for GATK is the issue. There are several alternatives in the tool group "NGS: Variant Analysis" at http://usegalaxy.org - including Freebayes. This has been discussed in other posts, for example:

Follow the 101 Tutorials here (includes an example using Freebayes): http://usegalaxy.org/galaxy101

Full 101 help is here: https://new.galaxyproject.org/learn/galaxy-ngs101/#non-diploid-variant-calling

ADD REPLY • link written 21 months ago by Jennifer Hillman Jackson ♦ 25k

Similar posts • Search »