A community effort to collect a curated set of analysis pipelines built using Nextflow.

nf-core

The issue with gzip/bgzip in consensus VCF has happened again (even though it makes no sense looking at the code), so I tried to investigate a bit. Sadly, the `test` profile is not too helpful, since it produces no consensus variants, and hence no VCF is created there. Callers were `strelka,freebayes` .

It can be easily fixed with a bit of `find` , but I'd rather avoid it because 1. it's annoying 2. I want the process to be as seamless as possible for future users

did youtry the consensus test profiles: <https://github.com/nf-core/sarek/blob/master/tests/postprocess_consensus.nf.test> `test` itself is very small and we use other profiles to test the various downstream combinations

I think I've found the issue, it's not the fault of Sarek, but of Vim. If you open a bgzip in Vim, it works just fine, but `:w` saves it as a normal gzip. Since `:wq` is muscle memory, I end up accidentally breaking VCFs when I inspect them to check integrity.