Proposal for a sublineage of BA.2.75
Earliest sequences: 2022-07-12 (India and Singapore)
Countries detected: India (2-4 from West Bengal, 2 from Telangana), Singapore (2), USA (1)

Mutations in addition to BA.2.75: ORF1b:V1706I and S:R346T.
ORF1b:V1706I is a large branch making up around 20% of BA.2.75. This lineage is not the only emergence of S:R346T on this branch - see below.

S:R346T is popping up repeatedly all over the BA.2.75 tree. Unfortunately most of the sequences are poor quality, with lots of NNNs and artefactual reversions, so it takes some work to figure out what's real and what's related to what; in particular, there are lots of sequences classified as BA.2.75 with S:346T that have S:339D rather than 339H and look like contamination with BA.2.74 or BA.2.76. But even the ones with S:339H still often get misplaced by Usher, and I had to edit the fastas in order to get them placed correctly.

However, I think there are at least 9 emergences of BA.2.75* + S:346T that look like they could be genuine. This lineage is the largest. The others are:

• S:346T directly on the BA.2.75 polytomy (3 sequences: EPI_ISL_14175634, EPI_ISL_14290506, EPI_ISL_14215529, from 3 different Indian states)
• S:346T directly on the BA.2.75.1 polytomy (2 sequences: EPI_ISL_14285583, EPI_ISL_14290481, one from Australia and one from India)
• 6 separate singlets. 2 of these are BA.2.75.1. Another 2 are on the ORF1b:V1706I branch, like the lineage proposed here; one of them is even misplaced within this lineage by Usher (see explanation below).

Here is an Usher tree that shows all of them: https://nextstrain.org/fetch/github.com/silcn/subtreeAuspice1/raw/main/auspice/subtreeAuspice1_genome_8f6f_12a0f0.json?c=userOrOld

While none of the S:346T lineages are big enough to need a designation yet, the fact that there are already so many of them is interesting in its own right, and that's why I'm proposing this lineage.

The reason why I've said "7-9 sequences" is that two of the sequences additionally have S:F486S and S:F490S. S:F486S is another mutation that's popped up a lot in BA.2.75 - I count at least 4 indepenent emergences, I expect I'll end up proposing one soon enough - and it's not clear whether these two sequences genuinely belong on the S:346T branch or whether S:486S came first. The 486S+490S branch could be deserve its own designation if it spreads enough.

In addition, a third sequence with S:F486S (WB-INSACOG-1931300701724, EPI_ISL_14303286) gets misplaced on the S:R346T branch by Usher, when it belongs instead on the branch defined by ORF1a:Q1198K. This singlet is omitted from the Usher tree below.

https://nextstrain.org/fetch/github.com/silcn/subtreeAuspice1/raw/main/auspice/subtreeAuspice1_genome_43005_1336c0.json?branchLabel=Spike%20mutations&c=gt-S_486&label=nuc%20mutations:G22599C

Genomes:
EPI_ISL_14049579
EPI_ISL_14101261
EPI_ISL_14175839
EPI_ISL_14215511
EPI_ISL_14215585
EPI_ISL_14295558
EPI_ISL_14302969
and possibly
EPI_ISL_14175700
EPI_ISL_14303178

Not going to try coming up with a cov-spectrum query yet, given how often 346T seems to be appearing...