Hiya. Without customisation of workflows in sarek, is there currently a route for having fastp trimming (for library adaptors) prior to passing these trimmed outputs to the UMI worklflow for extraction/grouping/consensus calling? Or having fastp extract UMIs prior to pushing to UMI workflow?

I implemented some extraction ability from fastp, see https://nf-co.re/sarek/3.8.1/docs/usage#how-to-handle-unique-molecular-identifiers-umis Basically UMIs are awkward because there are various places that they can be located.

Yeh, ok thanks.. But if you extract using fastp currently I don't think this can be passed (without workflow tweaking) to the grouping/consensus read building with fgbio tools. I would want to (best practises) trim adaptors etc. prior to doing consensus building . Just doing deduplication via GATK MD is possible, following extraction of UMI via fastp, but you're not then getting the most out of using UMIs (especially for high-coverage cancer pipelines) . Appreciate most can be done with use of

fastquorum

up front (before sarek), but I'm considering this for a clinical pipeline so version controlling two nfcore based pipelines (forked and customised anyway) is less desirable.

Yes, the current setup is fgbio before fastp. That could be changed if you are motivated enough.

Thanks. Yes will perhaps look at putting a second fastp option (and/or cutadapt to allow more control of 5' sequence trimming) inside the

fastq_create_umi_consensus_fgbio

workflow and associated options to run (or not) the current fastp task. .