[pag]

setting -m 2 is processing many more blocks. However, even at m=5 it was near-instant.

I noticed a section "What happens when my sample is a mixture, or diploid?" in the Quiver FAQ

Does this mean accuracy is suspect also for genomes with transposable elements?

[dalexander]

Quiver is jumping over the regions with inadequate effective coverage [== coverage after MapQV filter]. My understanding was you had close to 30x coverage?

You can try --minMapQV=0 --minCoverage=0 to completely disable input filtering, try it.
MapQV is effectively quantized at either 0 or 255 in my experience.

[pag]

yes, according to MIRA's mapping assembly (after pacbio error-correction), I have 30x coverage. There are a couple of problem regions of course (chiefly involving the transposons)

[dalexander]

Well, try turning off the mapQV filter and see what happens. The remaining explanation is that your draft assembly contigs are either redundant or ridiculously repetitive, so there is no confidence in the mapping.

[dalexander]

pag, if you are still having difficulties I can get you in touch with our tech support. Let me know.