Arthur Rand

Hello @lucagu1965, Sorry for the delay, but I think I've tracked down the problem and made a fix. You should not need this step: ```bash sort -k1,1 -k2,2n iPSC_MNCdLS1_5mC_allcontext.bedMethyl ```...

This change has been included in the latest release candidate. If you're still encountering a problem, please re-open this issue.

Hello @leetaiyi, Sure, that's not a bad idea. I agree that the docs should be as similar to what other tools to as possible to make it easier to compare.

Hello @K1999ban, > Does modkit pileup generate haplotagged bedmethyl files by default? If so, is there a way to disable this and generate a single, non-haplotagged file? Modkit does not...

Hello @bopohdr, > As i understood, if interested in only 5mC vs C (and not 5hmC), then collapsing is performed by splitting 5hmC probabilities equally between both states: collapsing 'h',...

@benjamin-bfeng Do you mean that the final `C` is aligned to a reference `CG` or that the read is missing a modification call when the _read sequence_ has a `CG`?...

@benjamin-bfeng, No problem. Could you paste a SAM record or attach a BAM file that is missing the modification call?

Hello @ziphra, Could you run `modkit modbam check-tags` [docs here](https://nanoporetech.github.io/modkit/intro_modbam_check_tags.html) on the CRAM or a small subset of it and post the output here? It will tabulate the errors encountered...

@ziphra It looks like these are different files you're using? `haplotagged.cram` vs. `19H01535.haplotagged.cram`? The first file looks to have almost entirely failed records, but the second one it is a...

Hello @ziphra, Ok. I don't think that having phased reads should matter, the phasing is just an tag in the record (same as the modified base probabilities). Would it be...