XBB with S:486P from US east coast - artefact or homoplasy? [9 sequences US, 1 AUS]

[corneliusroemer]

Looking for homoplasy of S:486P in XBB (due to the apparent fast growth of XBB.1.5), I noticed this apparent lineage from the US east coast that looks like it's very much not XBB.1.5 - it lacks all of the 3 non-S:486 mutations of XBB.1.5 and shares mutations with a few other sequences. And there's no obvious artefact I can see.

However it's not from a big branch of XBB. So I'm confused whether this is homoplasy or an artefact. Any thoughts welcome.

It's defined by XBB + 14178T,14362T,22792A,25539T

There's one sequence from Australia as well apparently.

https://next.nextstrain.org/fetch/genome-test.gi.ucsc.edu/trash/ct/subtreeAuspice2_genome_test_3737f_472530.json?branchLabel=back-mutations&gt=S.486L&label=id:node_7617106

EPI_ISL_16168600 EPI_ISL_16089826 EPI_ISL_16184328 EPI_ISL_16133568 EPI_ISL_15704652 EPI_ISL_16081991 EPI_ISL_15947141 EPI_ISL_16020835 EPI_ISL_16205389 EPI_ISL_16205459

[MCB6]

Maybe a little offtopic, but it's worth noting that in New England where most of the sequencing is done by Broad Inst with its unique pipeline, Pango is unable to call XBBs. So it assigns the putative XBB.1.5's to the generic BA.2 bin. But the trajectory of F486P growth is exactly the same as in the neighboring Connecticut and NY City where it's driven by XBB.1.5. (The final points on the graph are lower than they will eventually be because the most recent weeks are mostly represented by the earliest dates within the week)

[MCB6]

PS: in yesterday's upload from NYU Langone, half of the sequences had F486P, but a half of those were generic BA.2 too

[corneliusroemer]

This is a pangoLEARN/pangolin issue - if you use pangolin in Usher mode or Nextclade they should all get classified correctly.

[FedeGueli]

Hi @corneliusroemer spotted that in the last days and wondering the same of you. thx for proposing it.

[FedeGueli]

@corneliusroemer i think many more of this one popped up. i m checking them

[FedeGueli]

@corneliusroemer it reached 18 sequences and now looks legit to me: https://nextstrain.org/fetch/genome.ucsc.edu/trash/ct/subtreeAuspice1_genome_1751d_df0060.json?c=userOrOld&label=id:node_7695325

[corneliusroemer]

Very confusing, I'd probably wait a bit more in the hope things become clearer over time.

[AnonymousUserUse]

False milestone? It has been designated as XBB.6.1 if I understand correctly. @thomasppeacock @corneliusroemer