Chris Fields
Chris Fields
I traced the problem to this. Roles consuming roles can't define attributes; this is a well-known problem in Moose that isn't easy to fix. It has been TODO'd for now....
Blocking on an outstanding [Moose bug](https://rt.cpan.org/Public/Bug/Display.html?id=46347)
This might be worth checking in w/ our sysadmins on the cluster, my guess is their version of conda may not allow user-based installs. If so we'll likely need to...
@Juke34 we can address this but can't commit to a timeline. If you know the specific fix we also (gladly!) accept pull requests.
@benjjneb @terjekv I'd be happy to assist in testing this locally using Apple M1, and I believe we may have some access to arm64 nodes on our HPC. I guess...
@benjjneb sure, no worries!
Definite +1 on this; I've been using an alternative phasing and scaffolding tool (HapHiC) for our hifiasm assembly since it was potentially polyploid. It has worked pretty nicely using the...
At the moment, we're basically including them as part of the main reference genome FASTA and analyzing outside of methylseq. Having a way to include them separately would also be...
In the cases where there are controls added, we've performed what @mpiersonsmela but did a post-processing round to extract reads aligned to the control sequences, then evaluated methylation conversion on...
We've found the same if your sequencing facility is providing quality binned data from the Revio (default); in fact this seems to be generally better for any FASTQ binned data...