FW.1.1* branch with large deletions/frameshifts in Orf6/7ab/8 with Orf1a:K3353R and S:A684T (113 seqs)

[krosa1910]

First spotted by @NkRMnZr (Edited)

Gisaid Inquiry: A10323G,G23612A,C7119T
Earliest Sample: EPI_ISL_18248688 2023-09-03 England
Latest Sample: EPI_ISL_18688310 2023-12-18 England
It is mostly circulating in Great Britain, but samples in Denmark and US also
Tree of it
https://nextstrain.org/fetch/genome.ucsc.edu/trash/ct/singleSubtreeAuspice_genome_12107_1121d0.json?f_userOrOld=uploaded%20sample

Remember FW.1.1* gets K356T+R403K combo just like our dear JN.1*, yet it lacked F486P and it is proven that nearly all XBB without F486P does not survive well since XBB.1.5*.
Yet,

---

Edited by mod:
to fully appreaciate the relevance of this lineage better sharing also the complete tree of XBB.1.28:

More i think two relevant things are to be added :
First: XBB.1.28 in its mayority has a frameshift in orf6b described in a comment under the original proposal:

that the vast mayority (194/240 but further 19 miss it just cause sequencing issue, verified on Nextclade)) of XBB.1.28.1 sequences have a single nucleotide deletion del27208_27208 corresponding to the first nuc of the second codon of Orf6

This consequently causes a frameshift

This frameshift leads to a creation of a short new protein of just 10AA.
Its sequence is radically different from the first 11 A of the old Orf6:

OLD Orf6:1-11 .= MFHLVDFQVTI (ATGTTTCATCTCGTTGACTTTCAGGTTACTATA)
NEW Orf6:1-11 = MFISLTFRLL* (ATGTTCATCTCGTTGACTTTCAGGTTACTATAG)
..."

Second :this branch of FW.1.1 had the large deletion of Orf7ab and Orf8, described first by @ryhisner for GW.5.1.1, and found in FW.1.1 by @NkRMnZr back in October along with this sublineage.

[krosa1910]

I don't think this comes from contamination as S:A684T is not altogether a very popular thing, according to cov-spectrum it appeared 83 total within last 2 months, and 64 seemed to be contained in this branch. The growth though can be bit artificial due to the Britain sequencing technique, still it is something to look at.

[FedeGueli]

Thx @krosa1910 i ve edited the proposal a bit adding credits but especially with a little talk about this lineage:
XBB.1.28 as a whole started frameshifting Orf6 in a very short 11AA protein then had a Stop Codon at Orf7a:121 then got the Orf7ab orf8 deletion, so in my view it is the shortest genome circulating.

[FedeGueli]

CC @ryhisner @aviczhl2 @NkRMnZr @HynnSpylor @Over-There-Is @JosetteSchoenma
@krosa1910 :wait wait could this be a recombinant with GW.5.1.1?
cause i m noticing that this sublineage has not the orf6:2 frameshift that was in all XBB.1.28.1 and in the branch. So the only way i see it gained it back via recombination getting the large deletion too. Both lineages have circulated in England

The other thought i can have is that the large deltion messes up the sequencing and so the labs did base calling (WT) for the orf6 missreading,

cc @thomasppeacock @corneliusroemer @AngieHinrichs

[FedeGueli]

the recombinant hypothesis seems unlikely cause GW.5.1.1 has these mutations toward 3' end:
C28511T,T28723C,A28768G,G29810T
no one of them present in this FW.1.1.

[FedeGueli]

It is striking the fact that the T27205- (= Orf6:F2- > frameshift) deletion

is still there until the node T13728C:

But disappears completely within the node proposed here:

Tree:
https://nextstrain.org/fetch/genome-test.gi.ucsc.edu/trash/ct/subtreeAuspice1_genome_test_1ac7e_27e100.json?f_userOrOld=uploaded%20sample&label=id:node_3645960

As it can be seen from Covspectrum where only 30% of FW.1.1 + T13728C has Orf6:F2-):
https://cov-spectrum.org/explore/World/AllSamples/AllTimes/variants?nucMutations=T13728C&nextcladePangoLineage=FW.1.1*&nextcladePangoLineage1=GW.5.1.1*&

No easy explanation for it

[corneliusroemer]

Designated as FW.1.1.1 - I haven't dug into the deletion question