BA.2 sublineage with S:K147E, W152R, F157L, I210V, G257S, D339H, G446S, N460K, R493Q (73 seq as of 2022-06-29, mainly India)

[silcn]

Proposal for a sublineage of BA.2
Earliest sequence: 2022-06-02 (India)
Countries detected: India (5 seq, from 3 states)

Defining mutations:
S:K147E, W152R, F157L, I210V, G257S, D339H (mutated from G339D), G446S, N460K, R493Q (reversion)
ORF1a:S1221L, P1640S, N4060S
ORF1b:G662S
E:T11A

Don't think much needs to be said to explain why I'm proposing this. Very recent, long branch with 9 new spike mutations, detection in multiple states that aren't all close together (Maharashtra, Karnataka, Jammu and Kashmir). I expect quite a few people have been monitoring it :)

Usher tree is a bit messy because of some poor quality sequences, particularly the one from Jammu and Kashmir which has multiple artefactual reversions. As a result this lineage is placed on a branch with a couple of Indian sequences with reversions at S:954 (also probably erroneous - apparent reversions at this site seem to crop up a lot in Indian sequences). Despite what the tree shows, the evidence is currently consistent with all 9 S mutations having appeared on the same long branch. As usual, the 2nt mutation at S:339 is mislabelled.

https://nextstrain.org/fetch/github.com/silcn/subtreeAuspice1/raw/main/auspice/subtreeAuspice1_genome_42b02_161030.json?branchLabel=Spike%20mutations&c=gt-S_446&label=nuc%20mutations:C3796T,C3927T,C4586T,C5183T,A12444G,G22577C,G22898A,T22942G,G23040A,A26275G

Genomes:
EPI_ISL_13302209
EPI_ISL_13302252
EPI_ISL_13373059
EPI_ISL_13373170
EPI_ISL_13375776

Edit: cov-spectrum query https://cov-spectrum.org/explore/World/AllSamples/Past6M/variants?variantQuery=%5B6-of%3A+S%3A147E%2C+S%3A152R%2C+S%3A157L%2C+S%3A210V%2C+S%3A257S%2C+S%3A339H%2C+S%3A446S%2C+S%3A460K%2C+ORF1a%3A1221L%2C+ORF1a%3A1640S%2C+ORF1a%3A4060S%5D&
Missing some sequences that only have a month of collection

[ryhisner]

Man, we must be on the same wavelength, Silcn, because I noticed one sequence of this yesterday. I wasn't quite sure what to make of it—whether it was real or not, coming from India, as I don't really have the expertise to judge such things. Excellent spot, mate.

[silcn]

Now found in Germany: EPI_ISL_13378378, EPI_ISL_13378924
And also in Canada: EPI_ISL_13392500
EPI_ISL_13389935 is another Indian sequence, with NNNs covering the locations of all the S mutations but it shares all the non-S mutations.

[silcn]

More Indian sequences: EPI_ISL_13409385, EPI_ISL_13409444, EPI_ISL_13409465

[thomasppeacock]

Although this is still fairly limited in absolute sequence number, the divergent mutation profile, the wide geographical spread, and the rapidness that new sequences have emerged makes me think this should get designated pretty soon to faciliate its monitoring.

[c19850727]

@silcn most of these sequences have T5386G, and 2 of them also have A11537G though. I mean those brought up by @FedeGueli

[InfrPopGen]

Thanks for submitting. We've added lineage BA.2.75 with 3 newly designated sequences. Defining mutation(s) A22001G (S:K147E), T22016C (S:W152R).

[silcn]

@InfrPopGen the 4 sequences you've added to lineages.csv contain India/MH-INSACOG-CSIR-NEERI1939/2022 twice, which I presume is an error?

[InfrPopGen]

Thank you @silcn
I've deleted the duplicate line!

[silcn]

Among the 20 new BA.2.75 uploads from Maharashtra today is an apparent outlier sequence, EPI_ISL_13502528, which has S:147E, 210V, 257S, 339H, 493Q and ORF1a:1221L but is missing the other defining mutations and instead has some more of its own (none in Spike though). It is classified as BA.2 rather than BA.2.75 by Usher. If this spreads then it could potentially deserve a separate designation; I'll keep looking out for more.

Another of the new Maharashtra sequences (EPI_ISL_13502546) has S:681R and ORF8:27* - something else to keep an eye on...

[shay671]

As of this morning i found 38 samples.

https://nextstrain.org/fetch/genome.ucsc.edu/trash/ct/subtreeAuspice1_genome_40dd0_bf4e50.json?branchLabel=Spike%20mutations&c=userOrOld&label=nuc%20mutations:G23040A

[silcn]

@shay671 I count at least 46, 47 if you allow EPI_ISL_13502528.
NSP3_P822S, NSP8_N118S in GISAID is good at picking them up for now.

EPI_ISLs

EPI_ISL_13302209
EPI_ISL_13302252
EPI_ISL_13373059
EPI_ISL_13373170
EPI_ISL_13375776
EPI_ISL_13378378
EPI_ISL_13378924
EPI_ISL_13389935
EPI_ISL_13392500
EPI_ISL_13409385
EPI_ISL_13409444
EPI_ISL_13409465
EPI_ISL_13438623
EPI_ISL_13438754
EPI_ISL_13446524
EPI_ISL_13446529
EPI_ISL_13455147
EPI_ISL_13458019
EPI_ISL_13461861
EPI_ISL_13463939
EPI_ISL_13471039
EPI_ISL_13471048
EPI_ISL_13493348
EPI_ISL_13493438
EPI_ISL_13493615
EPI_ISL_13498391
EPI_ISL_13498432
EPI_ISL_13498452
EPI_ISL_13502529
EPI_ISL_13502534
EPI_ISL_13502536
EPI_ISL_13502537
EPI_ISL_13502538
EPI_ISL_13502544
EPI_ISL_13502545
EPI_ISL_13502546
EPI_ISL_13502548
EPI_ISL_13502550
EPI_ISL_13502552
EPI_ISL_13502554
EPI_ISL_13502555
EPI_ISL_13502559
EPI_ISL_13502567
EPI_ISL_13502568
EPI_ISL_13502569
EPI_ISL_13502571
EPI_ISL_13502576

[shay671]

You are 100% correct
All cluster together :
https://nextstrain.org/fetch/genome.ucsc.edu/trash/ct/subtreeAuspice1_genome_25dfc_c7fcf0.json?c=userOrOld&label=nuc%20mutations:C3796T,C3927T,A12444G,G15451A,A22190G,G22577C,G22898A,T22942G,A26275G

[thomasppeacock]

As this lineage is getting quite a bit of attention I've hidden a couple of comments earlier on about an unrelated recombinant to prevent confusion - please do feel free to open a new designation issue if the recombinant continues to grow

[silcn]

NSP3_P822S, NSP8_N118S in GISAID is good at picking them up for now.

Of course today we get some sequences missing NSP8_N118S... not sure there's a single GISAID query that captures everything anymore, but by my count 13 new BA.2.75 sequences were uploaded from India today.

[shay671]

Here is the case count by country/region for the 53 cases i found yesterday.

<style> </style>

India	Haryana	3
	Himachal Pradesh	3
	Jammu and Kashmir	1
	Karnataka	10
	Maharashtra	23
Germany	Baden-Wurttemberg	1
	Rhineland-Palatinate	1
UK	England	5
Canada	Alberta	1
	Ontario	1
Australia	Victoria	2
New Zeland	2

[FedeGueli]

With the new sequences on covspectrum although it has to be taken with a huge grain of salt, it shows growth advantage vs BA.5. @corneliusroemer @thomasppeacock @chrisruis
@AngieHinrichs
probably this has to be flagged to public health agencies

[karyakarte]

That seems to be true. The numbers of BA.5 are not increasing. BA.4 is on the verge of being extinct. BA.2.38 is almost same in numbers with the rest of BA.2. BA.2.74 and BA.2.75 were identified (amongst previous BA.2) when the Fasta were re-run on UShER today. @FedeGueli @silcn @corneliusroemer @thomasppeacock @chrisruis @AngieHinrichs

[FedeGueli]

@karyakarte consider that there are multiple sub lineages of potential nterest, designated, proposed or unproposed, emerged first and now circulating in India. I cannot weigh how much of the apparent growth advantage is due to these sublineages keeping BA.5 to lower levels than outside of India or due to intrinsic BA.2.75 advantage that cant be ruled out or affirmed with certainty yet.

[karyakarte]

@FedeGueli I totally agree. Sequencing coordinated by our center of 1698 isolates of May and June 2022 show BA.2 (48.62%), BA.2.12.1 (2.41%), BA.2.38 (39.33%), BA.4 (1.20%), BA.5 (3.11%) (with negligible numbers of other BA.2 sub-lineages making a total of 100%). This shows the reason for my earlier comment 10 hours back. that maybe some other lineage currently classified as BA. sub-lineage is growing. In our last run at BJGMC, Pune we found BA.2.74 and BA.2.75 that was red-flagged. Hence, we re-run Fasta available with us, with adequate coverage, at UShER. The results showed that a number of BA.2 turning out to be BA.2.74/BA.2.75. Hence, this issue is highlighted at appropriate fora with following comments, "Request to urgently re-run fasta on UShER again particularly samples from end of May and entire June.
The predominance of BA.2 on a waning wave was unexplainable. May be it is BA.2.74/BA.2.75 - with more than 80 mutations. A new designation for these variants seems essential, otherwise epidemiological response is blunted." We need to look into the matter @silcn @corneliusroemer @thomasppeacock @chrisruis @AngieHinrichs.

[FedeGueli]

Thx Prof. @karyakarte
if needed i can open in minutes an issue with all indian BA.2 sublineages we are tracking to help in a fast designation of them
cc @chrisruis @AngieHinrichs @corneliusroemer @thomasppeacock @InfrPopGen let me know. i have already everything tracked with route from route let me know if could be a good idea.

[FedeGueli]

@karyakarte here the link with all the BA.2 sub lineages we are monitoring in India:
https://docs.google.com/spreadsheets/d/1zKF3PKaosF3OCcW54gWkyV7Spz8uWrCsYef9IHbj7FA/edit?usp=sharing

[silcn]

Regarding BA.2 lineages in India: Japan just uploaded 42 sequences from travellers from India, sampled from 27 May to 16 June, which might give a more representative sample given the differences in sequencing quantity across India. Here is the lineage breakdown according to Usher:

17 BA.2.38 (of which 1 is #809)
15 BA.2 (of which 8 are #787, 2 have S:L452M but are not BA.2.56, and 1 has S:R346T+S:L452M but is not BA.2.74)
4 BA.2.56
2 BA.2.74
1 BA.2.75
1 miscellaneous BA.1/BA.2 recombinant
1 BA.4
1 BA.5.2

I agree with the comments in that thread that there is a very strong case for designating #787.

[karyakarte]

Thanks for the lightning-fast response! Your proposal for a fast designation to efficiently track emerging sub-lineages will answer the question of the rising case numbers by the same variant (BA.2) after waning of the 3rd wave it caused.

[FedeGueli]

@karyakarte i have updated that list with a prevalence region by region lineage by lineage sheet.
You can find it here

To me this sheet shows as it is actually a team work by BA.2 recently evolved fast sublineages in keeping BA.5 at the corner in India.
Probably this contributes to boost apparent growth advantages there to irrealistic levels.
And that if it is less fascinating for media headlines probably it is more worrying cause could mean BA.2 found multiple ways to compete vs BA.5 , my two cents maybe i am totally wrong.
cc @silcn @corneliusroemer @thomasppeacock

[karyakarte]

@FedeGueli thanks to you and the team's effort in trying to solve the puzzle of BA.2 equaling BA.2.38 in India. Numbers of BA.4 and BA.5 are also kept in check by both BA.2 and BA.2.38. But with many BA.2 turning out to BA.2.74, BA.2.75, now BA.2.76, and a few more to come, the puzzle of the same variant causing surge after a wave will be solved.
Further, as you mentioned multiple ways to compete, the strategy of BA.2 survival - mutational change in the immunodominant epitopes, seems to be successful. We are looking at it at present.

[silcn]

NSP3_P822S, NSP8_N118S in GISAID is good at picking them up for now.

Heads up for those tracking with GISAID: NSP3_S403L, NSP8_N118S is now a better proxy for BA.2.75 than the one above.

[shay671]

Two cases were spotted in Israel (I'll update after it's uploaded to GISAID); they are returning from France without known contact with someone from India. They are clustering with a sample previously collected in the Netherlands.
This is evidence for BA.2.75 already transmitting in Europe, not in relation to India.

https://nextstrain.org/fetch/genome.ucsc.edu/trash/ct/subtreeAuspice1_genome_3cd0c_403380.json?c=pango_lineage_usher&label=nuc%20mutations:C9693T

[RajLABN]

@AngieHinrichs @corneliusroemer @thomasppeacock @silcn
EPI_ISL_14402515-14402516, EPI_ISL_14402520 from Assam, India -- earliest BA.2.75?

https://nextstrain.org/fetch/genome.ucsc.edu/trash/ct/subtreeAuspice1_genome_dd44_876cf0.json?branchLabel=Spike%20mutations&c=gt-S_460,446,210&label=nuc%20mutations:T15867G

[silcn]

@RajLABN those are date errors.