JN.1 + 2-nuc ORF1b:V1271T (495 seq; Dec 29)

[ryhisner]

Transferred from sars-cov-2-variants/lineage-proposals#1143

Description
Sub-lineage of: JN.1
Earliest sequence: 2023-10-30, France – EPI_ISL_18510103
Most recent sequence: 2023-11-22, Israel – EPI_ISL_18572353
Continents circulating: Asia (38), Europe (5), Oceania (1), North America (1)
Countries circulating: Singapore (36), France (3), Israel (2), Switzerland (2), Australia (1), Canada (1)
Number of Sequences: 45
GISAID Nucleotide Query: C897A, G17278A, T17279C
CovSpectrum Query: Nextcladepangolineage:JN.1* & G17278A & T17279C
Substitutions on top of JN.1:
ORF1b: V1271T (NSP13_V348T)
Nucleotide: G17278A, T17279C

USHER Tree
https://nextstrain.org/fetch/raw.githubusercontent.com/ryhisner/jsons2/main/JN.1_G17278A_T17279C.json?c=gt-ORF1ab_5672&gmax=21555&gmin=266&label=id:node_4491220

Evidence
Two-nuc mutations are rare, but this one has emerged several times independently—most recent in FL.16.3, XBB.1.16, and JG.3—which might indicate strong selection pressure. Or perhaps it’s just some unusual replication error that randomly occurs now and then.

It’s possible this is actually two separate lineages. Five of the sequences have ORF7a:T28I, also found on another branch of JN.1. Usher joins these two together and give about 40 of the 45 sequences in this lineage an ORF7a:I28T reversion. Most likely ORF7a:T28I was acquired separately by the other branch and the five sequences in this branch. The other possibility is that the 2-nuc ORF1b:A1271T was acquired twice independently.

Curiously, two sequences on this branch have S:R683Q and one has S:P681S, both of which would abolish the FCS. Two others, from France, allegedly have S:R681P. Hard to take reversions like that seriously until you see it from multiple different labs, however.

Genomes

Genomes

EPI_ISL_18510103, EPI_ISL_18524417, EPI_ISL_18541866, EPI_ISL_18541869, EPI_ISL_18544451, EPI_ISL_18544456, EPI_ISL_18544544, EPI_ISL_18544568, EPI_ISL_18544576, EPI_ISL_18544579, EPI_ISL_18544588, EPI_ISL_18544592, EPI_ISL_18544610, EPI_ISL_18544689, EPI_ISL_18544691, EPI_ISL_18544700-18544701, EPI_ISL_18544719, EPI_ISL_18544726-18544727, EPI_ISL_18544735, EPI_ISL_18544754, EPI_ISL_18544763, EPI_ISL_18544773, EPI_ISL_18544776, EPI_ISL_18544781, EPI_ISL_18544783, EPI_ISL_18544794, EPI_ISL_18544804-18544805, EPI_ISL_18544812, EPI_ISL_18544819-18544820, EPI_ISL_18544836, EPI_ISL_18544842, EPI_ISL_18544857, EPI_ISL_18544872, EPI_ISL_18544880-18544881, EPI_ISL_18544884, EPI_ISL_18551379, EPI_ISL_18553476, EPI_ISL_18569059, EPI_ISL_18572344, EPI_ISL_18572353

[FedeGueli]

This is also high in Malaysia 25% of samples there

[AngieHinrichs]

Usher joins these two together and give about 40 of the 45 sequences in this lineage an ORF7a:I28T reversion. Most likely ORF7a:T28I was acquired separately by the other branch and the five sequences in this branch.

Yeah, that would be my guess too. I will try a little prune/re-opt/re-place to see if I can at least get rid of the flip-flop on 27476 (ORF7a:28).

[FedeGueli]

5% of samples in Indonesia in Decmber . now 378

[alurqu]

If I'm correct, G17278A T17279C yields an RNA sequence of AAACGAAU which differs from the canonical sarbecoronavirus TRS-B in only the last nucleotide being U instead of C. Could this function as a new suboptimal TRS-B?

If so, the next potential start codon I see would be at out-of-frame starting at 17300 but running not far with a stop codon in that frame starting at 17315 yielding only MSFVL. Unless that is some high-value signaling molecule, it would seem a waste of energy for the virus to produce an sgmRNA for this. So perhaps this is not significant or the trailing U reduces the sgmRNA transcription cost.

[ryhisner]

Oh, dang, you're right. Nice spot! Not sure how I missed that one, but it's definitely just one nuc away from the optimal TRS. I think the final C is pretty essential, so I don't think this would result in the transcription of much of anything. But it could be a clue that this 2-nuc mutation formed through some sort of recombination with the TRS-L. Normally, you'd expect to see more than 7 matching nucs or at least some pretty good extended homology, but not always.

[FedeGueli]

great catch @alurqu . btw this is 390 now with the last 47 samples coming from outside Singapore and from 9 different countries. Now with these additional data @corneliusroemer i would not wait to designate it.

[ryhisner]

There are, uh...102 new sequences of this lineage today from Singapore out of 495 sequences uploaded from there today, so over 20%.

[FedeGueli]

There are, uh...102 new sequences of this lineage today from Singapore out of 495 sequences uploaded from there today, so over 20%.

Yeah i have to say that means it is pretty stable cause it was floating around 20 25 if i am not wrong, last time i checked.

rechecked it is increasing from 15% to 20% so it is growing faster than rest of JN.1 EDITED

[xz-keg]

56/242=23% in the new singapore batch, up from 21% 2 weeks ago.

[corneliusroemer]

Stretches of As (or any nucleotide) are always prone to sequencing errors as well, that could in general explain homoplasy as well. Not saying that's the case here though - especially given it's reliably called in a number of countries.