Brent Pedersen

@empathy-c as I mentioned in another issue, you won't be able to use python 3.10 with version 0.20.0, you'll need an older python version.

Hi @empathy-c sorry for the troubles, indeed we're having difficulty with the builds. I would suggest to use bioconda to get that version.

Hi, you are using an old version of cyvcf2 on a new version of python. So you'll have to use an old version of python to do this. Also note...

Hi, sorry for the delay. Is this still an issue? We are using github workflows to build: https://github.com/brentp/cyvcf2/actions/workflows/wheels.yml Do you have some guidance on changes? I assume these have been...

closed by #262

Thanks for the reply. Re score-only positions, that's great! I can access the struct directly. For extension, an API for that would be great. Meanwhile, I don't understand how to...

not sure how this: ``` File "/home/ryang/miniconda/envs/balrog/lib/python3.6/site-packages/bwameth-0.2.2-py3.6.egg-info/scripts/bwameth.py", line 359, in handle_header sq, sn, ln = toks # @SQ SN:fchr11 LN:122082543 ValueError: too many values to unpack (expected 3) ``` is...

500 is a good value. You definitely don't want to use the full chromosome length. It uses something like a sliding window. Hope that helps.

I think this is a limitation of the implementation here. If you need p-values that low then you must use scipy. I thought recent versions of scipy were quite fast.

Thanks for going through this. I'd prefer to stick to nim < 2 for this as I haven't tested fully on nim 2.0+. And the OSX build would be nice...