Cornelius Roemer

You're right @tsibley, I assumed it was pulling from Github releases, but it isn't. Recipe is set up to pull from PyPI 👼

You'd have to build from source - conda forge doesn't like repackaging binaries. But maybe they are more lenient with ARM?

You're right, could add the other Omicrons. But then they're so rare, most B.1.1.529 are actually BA.1 just bad sequences. Current situation mostly down to laziness on my side, 1%...

Oh that's amazing, I had thought about doing something like that but time... If you want to submit a PR happy to merge. Some thoughts: I would drop BA.3 since...

Wouldn't include BA.1.1 separately It'd be a mess because it's not independent but a descendant of BA.1. BA.1 and BA.2 for themselves and compared is what's most interesting I think...

Adding weights to the regression would be really really nice to have. 👍

Random name is perfectly fine in this case, what else should it be :) could of course be passed as a command line argument as extra feature as add on.

Design improvements address the problem of the table being chopped off after 800px See https://github.com/nextstrain/sphinx-theme/pull/18?notification_referrer_id=NT_kwDOAX_wQbMzMjgyNTE1NjUyOjI1MTYxNzkz Now we can add extra columns for `type` and `example` @chaoran-chen

Fair point. I just wouldn't apply software log levels in this case. Warning should be reserved for a real problem. Not for QC issues. Failure to align is basically because...

Agree with most here. Maybe adding context/explanation would be all that's needed. The biggest issue is I think when users install Nextclade, they run Nextclade for the first time with...