Update clade definition and coloring #274
Conversation
This commit: 1. Very slightly changes clade definition for "clade II". This results in no change to the resulting clade assignments. Previously clade II definitions included nucleotide positions 35253 and 150970. However, neither of these positions show variation: - nextstrain.org/mpox/all-clades?c=gt-nuc_35253 - nextstrain.org/mpox/all-clades?c=gt-nuc_150970 I updated these sites to 20605 and 174146 which show variation between clades I and II. 2. The color ordering lacked entries for clade Ia and clade Ib. This resulted in gray colors for these two clades at nextstrain.org/mpox/all-clades.
There was a problem hiding this comment.
I can't speak to the previous clade-II mutations, but the new ones are unique to clade-II in a just-built mpxv tree. Note that there are no tips which are labelled as clade-II or clade-I (only internal nodes) which has a few oddities in auspice (see nextstrain/auspice#1275) - they appear as distinct colours but not in the legend, you can't filter to them directly etc.
"The color ordering lacked entries for clade Ia and clade Ib" (in the mpxv build) is #272 which is fixed here (and the clade-I build also gets non-default colours as a result).
There was a problem hiding this comment.
Thanks for double checking the clades TSV.
@trvrb I think you might have made a digit swap when hand-typing 35352/35253 - it's 35352 in the clades.tsv and your PR mentions 35253 (an anagram with the 3rd and 5th digit swapped).
What used to be in the clades.tsv has diversity at that site:
https://nextstrain.org/mpox/all-clades?c=gt-nuc_35352&gmax=36253&gmin=34253
But given it's only in Ia not I it's best to remove/replace in any case.
150970 might have shifted to 150968 (which does show I/II diversity A/T due to changes in Nextclade alignment - good spot.
This PR:
Previously clade II definitions included nucleotide positions 35253 and 150970. However, neither of these positions show variation:
I updated these sites to 20605 and 174146 which show variation between clades I and II.
The resulting
all-cladesbuild looks like:I don't think this should be too controversial, but could either @jameshadfield or @corneliusroemer take a very quick look?