Michael Hall

> I have no idea why the error rate there should be different as the error rate used for alignment things (although I would hesitate to say that they should...

I have been playing around with these two error rates today. The data I am parameter sweeping with is Nanopore Mtb data. They have "truth" assemblies from PacBio CCS and...

> what are the default values for e and E? e = 0.11 (nanopore) and E = 0.01 (always) I'll put together those other plots today

> Is it possible to replot these only for samples with >40x depth, so we can disentangle coverage? Here are the plots with **samples that have coverage over 40x** (which...

Leandro and I were speaking about this before. Basically, the thing that worries me is that say you sketch a read and only one of it's kmers hit somewhere. Then...

Yeah, I agree about size selection. I wouldn't advocate for selecting reads based purely on size, but maybe some kind of metric that takes into account read length, hits, (maybe...

I want to dig this up again. Based on discussions which I don't think have been documented here, I think it might be best if we remove the `--max_covg` parameter...

Well, we'll need to remove this parameter and it's associated functionality in the code before closing.

I think outputting only genotyped alleles by default would make a lot of sense. I am currently always normalising the pandora VCFs to get around this. We could then have...

In the process of working on this issue we can also probably fix #118