BA.2 + S:ins212TVGG in England region (800 genomes as of 2022-05-08)

[54mu]

Description

Sub-lineage of BA.2
labeled as Spike_ins213VGGG on GISAID
earliest collection date: 2022-01-05
latest collection date: 2022-04-19
Circulating in England, UK

Complete list of districts counts:

division	virus
England	298
Scotland	7
Baden-Wurttemberg	3
Galway	2
Massachusetts	2
Bavaria	1
Borgerhout	1
California	1
Hovedstaden	1
Namur	1
Nebraska	1
North Rhine-Westphalia	1
Wales	1

Proportion over the sequenced BA.2 genomes in England (rolling average of 7 days, 95% confidence interval):

currently assigned in GISAID to 'BA.2', 'BA.2.5', 'BA.2.9', 'BA.2.16', 'BA.2.3', 'BA.2.10'

Genomes

see this gist for the list of GISAID genomes

Evidence

We described this cluster in https://doi.org/10.1101/2022.04.11.487924, together with the one proposed in #583

Phylogenetic tree (updated to 2022-03-27):
https://nextstrain.org/community/54mu/SARS-CoV-2_BA.2_RIR1insertions/treeViz?c=BA_group
the proposed lineage is here marked as BA.2+ins(LIII) as referenced in our preprint. BA.2+ins(L) marks the lineage proposed in #583
for extended methods see here

[corneliusroemer]

Thanks for proposing this. I did an independent survey of insertions and also identified this as the second biggest BA.2 insertion in terms of sequences. So that's reassuring :)

In nucleotide coordinates this insertion is: 22198:insACAGTAGGAGGA

With GISAID data up to around the 25th of April I counted 345 sequences.

Would be good to find nucleotide mutations that are shared by (nearly) 100% of these sequences to distinguish from other BA.2 and also some big branches with nuc mutations within it (so less than 100% but significant numbers).

PangoLEARN, UShER and Nextclade-Pango wouldn't be able to work off an insertion alone (at this point) unfortunately.

@54mu great you have a tree! It would be more useful if it was annotated with nucleotide mutations that are on different branches (it should be possible to reconstruct those with augur ancestral) - basically running a more complete version of the Augur ncov pipeline.

[corneliusroemer]

Turns out we are lucky! This insertion appeared on top of nuc mutations 4202T followed by 25253T, the latter one being equivalent to S:T1231S [confirmed by dropping all sequences with this insertion into Nextclade]

The query for the combination of nuc mutations alone results in 654 sequences, of which at least half have that insertion. Some of the missing ones could miss due to bioinformatic errors.

So I think we should be able to easily designate this one as BA.2* + 4202T, 25253T rather than working with the insertion. @chrisruis

https://cov-spectrum.org/explore/Europe/AllSamples/Past6M/variants?nucMutations=4202T%2C25253T&pangoLineage=BA.2*&pangoLineage1=BA.4&

One S2 mutation plus a 4 AA insertion in NTD of S1 makes an interesting lineage.

The branch with the insertion also seems to have a slight growth advantage on the order of a few percent per day.

[54mu]

It would be more useful if it was annotated with nucleotide mutations that are on different branches (it should be possible to reconstruct those with augur ancestral) - basically running a more complete version of the Augur ncov pipeline.

Hi, thank you for the tip, I updated the linked tree.

May I ask how did you obtain the nucleotide substitutions? I ran a chi squared test on substitutions inferred from the alignment used for the tree and found the exact same substitutions, plus two that do not appear in cov-spectrum (I may need to recheck my script).

[InfrPopGen]

Thanks for submitting. We've added lineage BA.2.41 with 561 newly designated sequences, and 51 updated designations from BA.2. Defining mutation(s) A25253T (S:T1231S).