Option for pacbio CCS/HiFi reads

[Kirk3gaard]

Hi Ryan
Great job. Looks like yet another really cool tool solving a bunch of important problems with long reads!

I noticed that you have option for pacbio reads. However, it might be great to have an option for CCS/HiFi reads also.

Have you tried running it on graphs from CANU? Your circularisation step would be great there too at removing overlapping ends I assume.

Best regards
Rasmus
(I can finally skip my dirty hacks for showing circular contigs from miniasm nicely in bandage)

[rrwick]

What exactly would need to change for CCS reads? Just the minimap2 preset? Sorry - I have essentially no experience working with them!

And regarding Canu graphs, no it won't work. Minipolish assumes it's being run on a miniasm graph, so it will complain and quit if you give it a Canu graph. But Canu does its own consensus step, so I'm not sure if running Racon on its graphs would be necessary.

[Kirk3gaard]

Yeah I think Heng Li has thought this through
minimap2 -ax asm20 ref.fa pacbio-ccs.fq.gz > aln.sam # PacBio CCS genomic reads
(https://github.com/lh3/minimap2)
Me neither but I assume with the recent upgrades in throughput it might be gaining popularity.

Okay I just noticed that they removed the graph file in v. 1.9 anyway.

[Kirk3gaard]

Maybe just add the full suite of minimap2 options? It would be nice to be able to run minipolish with short reads as well :D.

[jelber2]

See this pull request to add PacBio CCS/HiFi option to Minipolish #7