Proposal for a Sublineage of BA.5.6 with S:K444T [238 seq on Usher, 229 seq on GISAID, Aug 15]

[ryhisner]

Description
Sub-lineage of: BA.5.6
Earliest sequence: 2022-5-27, USA, New Jersey — EPI_ISL_13544819
Most recent sequence: 2022-7-22, USA, New York — EPI_ISL_14178050
Countries circulating: Primarily USA and Belgium: USA (86), Belgium (26), Mexico (7), Ecuador (6), France (2), UK (2), Canada (2), Chile (1), Luxembourg (1), Netherlands (1), Peru (1), Puerto Rico (1) —as found using GISAID query “Spike_K444T, NSP13_A389V”
Number of Sequences: 122 on Usher Tree; 136 using GISAID query “Spike_K444T, NSP13_A389V”
Substitutions on top of BA.5.6:
Spike: K444T
Nucleotide: C8605T (synonymous), A22893C

USHER Tree
https://nextstrain.org/fetch/raw.githubusercontent.com/ryhisner/jsons/main/subtreeAuspice1_genome_34ea_9274e0-BA.5.6%2BK444T.json

Evidence
CovSpectrum: 26% (CI:9%-43%) weekly growth advantage relative to nextcladePangoLineage:BA.5.6* in the last three months in USA:
https://cov-spectrum.org/explore/United%20States/AllSamples/Past3M/variants?variantQuery=Nextcladepangolineage%3ABA.5.6*&variantQuery1=Nextcladepangolineage%3ABA.5.6*+%26+S%3AK444T&analysisMode=CompareToBaseline&

CovSpectrum: 43% (CI:4%-82%) weekly growth advantage compared to nextcladePangoLineage:BA.5.6* in the last three months in Belgium:
https://cov-spectrum.org/explore/Belgium/AllSamples/Past3M/variants?variantQuery=Nextcladepangolineage%3ABA.5.6*&variantQuery1=Nextcladepangolineage%3ABA.5.6*+%26+S%3AK444T&analysisMode=CompareToBaseline&

According to the Bloom Lab RBD Escape Calculator, S:K444 is the prominent escape site for antibodies known to neutralize BA.4/5.
https://jbloomlab.github.io/SARS2_RBD_Ab_escape_maps/escape-calc/

S:K444 mutations likely impose some cost in terms of ACE2 binding and RBD expression. Here’s a comparison of the reduction in ACE2 binding and RBD expression of the three most common K444 mutations seen thus far (on a BA.2 background), according to the Bloom Lab RBD Heat Map. K444T imposes a lower cost than K444N in terms of reduction of ACE2 affinity (-0.10 vs -0.16) but a higher cost in loss of RBD expression (-0.19 vs -0.08). K444T has the same reduction as K444R in ACE 2 affinity (-0.10) but a much higher reduction in RBD expression (-0.19 vs -0.04).
https://jbloomlab.github.io/SARS-CoV-2-RBD_DMS_Omicron/RBD-heatmaps/

Genomes
GISAID_hcov-19_ids_2022_08_02_14_06.csv

[FedeGueli]

I was tracking this since you first spotted it and it is keeping growing without the usual cluster effect (All up and then all down)
Comparing it with the new global dominant lineage BA.5.2* shows how it seems to keep an advantage versus BA.5.2* too:

https://cov-spectrum.org/explore/United%20States/AllSamples/Past3M/variants?variantQuery=Nextcladepangolineage%3ABA.5.2*&variantQuery1=Nextcladepangolineage%3ABA.5.6*+%26+S%3AK444T&analysisMode=CompareToBaseline&

Data from different states are still sparse . Probably we need a week or two to see how it will compete nationwide

[corneliusroemer]

This is up to 200 sequences now on GISAID and clearly growing faster than BA.5(.6) - given the Spike mutation with escape potential this deserves immediate designation in my view.

It also makes up ~5% of recent sequences from Ecuador (though only 7 sequences in total, still interesting)

[InfrPopGen]

Thanks for submitting. We've added lineage BA.5.6.2 with 45 newly designated sequences, and 24 updated designations from BA.5.6. Defining mutation(s) A22893C (S:K444T).