VQSR not fully implemented about dna-seq-gatk-variant-calling HOT 4 OPEN

Well spotted and thanks for documenting this in so much detail, with good suggestions. You and anybody using this pipeline is welcome to implement this via a PR, and we will look into implementing this if we can find some time. But we are not actively using this pipeline ourselves, so this might be a while.

For further inspiration, the base quality score recalibration (BQSR) setup might also be a good place to look, especially as this uses some of the respective files available in this workflow:

from dna-seq-gatk-variant-calling.

Thanks! I am also not actively using this pipeline here, as I'm developing my own that already fixes the issue, as stated above, so I won't have much time to work in this either here :-(

I also noted the BQSR setup has some similarity in GATK - although it seems that the wrappers for that changed. In previous versions, both steps of BQSR were combined into one wrapper, and only split into two later on. That might have been the reason why this was missed in the pipeline here.

from dna-seq-gatk-variant-calling.

Sure, that could be a root cause here. Or simply overlooking some details...

Does your workflow above follow the GATK best practices and do you actively maintain it? It might be worth deprecating this workflow in favor of yours, if yours ticks the right boxes...

Also, I realized yours isn't yet listed in the Snakemake Workflow Catalog. That catalog is basically a nightly GitHub crawler that aggregates all snakemake workflows that adhere either to its inclusion criteria. And if you walk the couple of extra meters to enable standardized usage, you even get a quick and easy deployment help for you workflow that you can for example cite in its README.md. As a good example, see the usage info for our dna-seq-varlociraptor workflow.

from dna-seq-gatk-variant-calling.

Does your workflow above follow the GATK best practices and do you actively maintain it?

As far as I am aware, it does indeed implement the best practices. And yes, at least for a while I plan to maintain it. We are currently working on a publication describing the pipeline as well.

Also, I realized yours isn't yet listed in the Snakemake Workflow Catalog.

Indeed, I'd love to have it listed in the catalog! I'll hopefully find some time to walk these extra meters soon ;-)

from dna-seq-gatk-variant-calling.