Added dragen alignment pipeline version 4.2.4

[alexiswl]

TODO

• Wait for Added cttso coupling workflow to production #382

[victorskl]

Alexis; confirming -- is this new dragen alignment pipeline is alignment QC pipeline for both WGS and WTS, correct..? And that will be replaced wgs_alignment_qc with Ocicat release as discussed..?

[alexiswl]

The WGS QC Pipeline has one step (which calls the dragen alignment pipeline).

So I think we can shrink this down and just call the dragen alignment pipeline instead, we will use this for both WGS and WTS QC steps with the only difference being the '--enable-rna' parameter

[alexiswl]

Thoughts on this @skanwal, means a change to the WFL ID in umccr/infrastructure#334

[victorskl]

Portal Release - Ocicat: Run WTS QC workflow through Portal before running Dragen WTS Tumor Only Workflow

[skanwal]

I second this idea @alexiswl. Initially, I believe, we created wgs-qc pipeline thinking we might need to add additional steps to it down the road specifically for WGS data.

Did we test dragen alignment pipeline for WTS data in dev?

[skanwal]

LGTM. Assuming it has been tested for WTS data in dev?

[alexiswl]

Did we test dragen alignment pipeline for WTS data in dev?

Pretty sure we ran the wgs qc for WTS as a test. Can run alignment for both WGS and WTS this afternoon

[alexiswl]

Done, with

WTS

wfr.82f89531a0b44d26b0e717c2f5b7178f

WGS

wfr.a241f7e6ec06487db6a9eecc8f694470

4.2.4 didn't have the --enable-map-align-output parameter available so have added that in.

It is a redundant parameter since we are not doing variant calling, but good to have regardless.

[alexiswl]

Also added to production_workflows project

[alexiswl]

Investigating missing RNA Quant Metrics and Transcript Coverage from multiqc in wfr.82f89531a0b44d26b0e717c2f5b7178f outputs.

May be that just adding --enable-rna=True isn't enough.

Also note that the output files for dragen still contain wgs_* prefixes :/

[alexiswl]

Ah need the --enable-rna-quantification parameter for transcript and quant metrics. This also requires an annotation file to be loaded. Are we happy to have these omitted? This 'QC' workflow can then just be with respect to alignment quality (and depth).

[skanwal]

--enable-rna-quantification parameter for transcript and quant metrics. This also requires an annotation file to be loaded.

I'd suggest we should add these as looking at the QC report we are also missing rRNA%, which might be a useful metrics for the lab.

Also note that the output files for dragen still contain wgs_* prefixes :/

As also mentioned yesterday - not surprised..

[alexiswl]

I'd suggest we should add these as looking at the QC report we are also missing rRNA%, which might be a useful metrics for the lab.

We can add in --rrna-filter-enable=true then too -
from dragen v4.2 docs

[alexiswl]

Reran with wfr.9fca2a78578741c1b3b04fb540ff9162.

• %rRNA ✅
• Transcript Coverage Plot ✅
• Quant Metrics Plot ❌

Tried running with multiqc 1.15 (released a few days ago) but I forgot to commit the search patterns file so that's a new PR here MultiQC/MultiQC#1969.

tl;dr, we can create our own custom container if we really need the quant metrics in multiqc?

[skanwal]

tl;dr, we can create our own custom container if we really need the quant metrics in multiqc?

For dragen-transcriptome workflow, we'll need the plot. So, it'll be good to have it added.
Thanks for the upstream PR.

[alexiswl]

This is using from this commit MultiQC/MultiQC@eb98188

Copied and created an image under https://quay.io/repository/umccr/multiqc?tab=tags

Output here -

PTC_NebRNA210629_dragen_alignment_multiqc (1).zip

Have then updated the dragen alignment workflow to use new multiqc tool (have added this in as a separate PR #384)

[alexiswl]

Waiting for #384

[alexiswl]

@skanwal please let me know if this is all okay from your end?

Would like to get stg, prod PRs done for WGS/WTS QC infrastructure and fix up umccr/data-portal-apis#611 with the additional parameters discussed above.

[skanwal]

All good from my end @alexiswl.