Our latest study is live on @medrxivpreprint


We asked why some (unvaccinated) individuals remain asymptomatic after SARS-CoV-2 infection, using 3 large cohorts.

And learned a lot on HLA & role in chronic phase of infection vs. acute phase

Check 🧵medrxiv.org/content/10.110… Not everyone responds the same to SARS-CoV-2 infection, even before vaccination.

One of the clearest example is the UK SARS-CoV-2 Human Challenge Characterisation study.

Only 1 of 18 infected was asymptomatic. Severity of infection is like a quantitative trait.

Curious about Delta-Omicron recombinants?

Our study 👏 @shishiluo @my_helix: medrxiv.org/content/10.110…

Co-circultation of both variants
20 samples with co-infections
Evidence of a few recombinations in co-infections
& 2 infections with 100% recombinant

Details in 🧵 @shishiluo @my_helix 2/
Recombination is one way a virus can evolve. It allows to 'test' a new combination of mutations, which could provide an advantage or not.

A great review on how RNA viruses recombine: nature.com/articles/nrmic… by @edwardcholmes @SimonLoriereLab

For SARS-CoV-2, should be 👇

BA.2 situation in 🇺🇸

Latest @my_helix sequences (from today): BA.2 was 9 of 780 in samples from Feb 01 - 04.

- BA.2 seems to be ~1% in USA for now.
- BA.1.1 at ~75%

public.tableau.com/app/profile/he…

and in other countries? 2/
BA.2 is dominant in:
India, Denmark, Sweden, Singapore, Austria, South Africa & Philippines

Check out the awesome graphs on covariants.org/per-country

by @firefoxx66 and her team

1/

A couple more days of data from @my_helix surveillance:

BA.2 back up at ~1% on Jan 29 & Jan 30

Still growing much slower than what expected, but maybe a sign that it is not completely at bay

Jan 29:
1 Delta
162 BA.1
465 BA.1.1 (73%)
7 BA.2 2/

So far, it does not look like any state in the US is particularly enriched for BA.2. Numbers correlate with sampling by state.

See 👇

source: public.tableau.com/app/profile/he…

Great talk from @ETCirulli at @DukeCAGPM this morning on

her research with @my_helix @HealthyNV @grzymski & @uk_biobank

to identify the best genes and variant models to include when doing genetic screens in the entire population

🧵

https://twitter.com/DukeCAGPM/status/1491415547862749189

2/
Screening an entire population is different than doing the exome to identify cause of a rare disorder.

Why does 1-3% of people develop very severe COVID-19 (before vaccine)?

Many studies showed the importance of type 1 IFNs, starting with human genetics see 🧵)

Reviewed by @tutu_ZQ @casanova_lab & COVIDHGE consortium

nature.com/articles/s4158…

1/9 Human genetics often allows to show causation. For severe COVID-19, the first strong hint was that:

- a couple young patients with critical COVID-19 had autosomal recessive deficiency of IRF7 or IFNAR1

- experiments showed that their cells could not control SARS-CoV-2

2/9

Latest @my_helix sequences live: public.tableau.com/app/profile/he…

Almost all Delta (B.1.617.2 & AY.)

- 1,319 of 1,380 week 30
- 146 of 149 week 31

Delta clearly displaced Alpha (B.1.1.7) in US & outcompeting Gamma (P.1), Lambda (C.37) & others.

What about B.1.621? 2/
B.1.621 & B.1.621.1 don't have a @WHO greek letter, but they are interesting

because of mutations in spike:
E484K, N501Y

& P681H that affects same amino acid as
P681R -> one of the mutation thought to cause great transmissibility of Delta
see nature.com/articles/d4158…

Update on Alpha, Gamma & Delta. Check our updated preprint: medrxiv.org/content/10.110…

- Alpha now < 3% in US
- Delta taking over. incl. outcompeting Gamma.
- Analysis of Cq levels suggests Delta infections lead to higher viral loads.

Details in 🧵 2/
We looked at 370,000 positive tests from US in 2021.
- tests from 50 states, but really it is 10 states that are well covered, esp. FL.
- most samples from a national retail pharmacy.
- our test are RT-qPCR.
- we also sequenced 56,000 positives & assigned Pango lineages.

New @my_helix viral sequences.

Up to July 8 (collection date).

- Delta now >80% of sequences in US.
- Alpha at 5% and decreasing.

And main result: Gamma also starting to decrease (more sequences needed to be sure) @my_helix 2/
More details:
- seqs not yet on GISAID
- numbers for the last 6 weeks are below

Based @my_helix test & seq data,

Delta (B.1.617.2) is now the most prevalent lineage in US (or where Helix tests)

- June 26, Alpha (B.1.1.7) < 20% of positives. See 👇 Trend 📉
- 10 days ago, Delta represented 30% of sequences, & 44.6% of S-positives. Trend 📈

links in 🧵 @my_helix To see the SGTF results, check: public.tableau.com/app/profile/he…

In May & June 2021, 99.3% of SGTF we sequenced were Alpha (B.1.1.7) making it a great proxy to look at very recent (& also larger N) for Alpha

06/25: Alpha ~22% (of 202 positives)
06/26: Alpha ~18% (of 182 positives)

What is happening with Alpha, Delta & Gamma in the US?

Our current analysis is now on @medrxivpreprint

medrxiv.org/content/10.110…

Using @my_helix testing & sequencing data, we see that:
- Alpha is rapidly 📉
- Delta & Gamma are 📈

Keep reading for nuances 2/
Based on data that Delta took over in India, England (see @PHE_uk) & other countries, we hypothesized that same would happen in 🇺🇸

Despite fact that health policies & viral background are different.
- Alpha was >90% of cases in England.
- Alpha plateaued at ~70% in US.

When will B.1.617.2 (Delta) be the dominant variant in the US?

Could be next week (or next 2 weeks) based on trends in our testing data & sequences available.

SGTF (~B.1.1.7) now < 50% based on last 5 days.

source: @my_helix @DiseaseOutbreak P.1 (Gamma)
B.1.526/.1 (Iota)
B.1.617.2 (Delta)
are the other main variants

Based on recent small numbers (check outbreak.info), Delta is trending up, but Gamma and Iota are not increasing anymore.

More sequences will provide more accurate info.

We will aim to update @my_helix dashboards even more frequently

SGTF updated with stats up to June 7
public.tableau.com/app/profile/he…

Info from UK + early stats on @GISAID & outbreak.info point to Delta driving part (most) of recent📉 in SGTF in US. But seq data is lagging. To read more on B.1.617.2 and its rise in UK: assets.publishing.service.gov.uk/government/upl…

- B.1.617.2 overtook B.1.1.7 recently. Observed by sequencing, & also by looking at SGTF data 👇

Something to keep in mind when looking at SGTF in US

- Also US had similar path to UK for B.1.1.7

We updated @my_helix viral surveillance dashboards.

- SGTF dashboard still here => important to see in real time if a variant is outcompeting B.1.1.7 in some states

- Selected variants dashboard. You can see P.1, B.1.617.2 & others. public.tableau.com/profile/helix6… 2/
Using the SGTF dashboard:
public.tableau.com/profile/helix6…

In some states like FL, SGTF 5-day avg is 📉 66% on May 9 compared to 71% 10 days before. & SGTF likely still >95% B.1.1.7

So what is the 34% and maybe growing?

Great thread from @trvrb on growth of P.1 in the US.

We can see it in @my_helix data. See this dashboard: public.tableau.com/profile/helix6…

In all US, we see ~20% of non-SGTF tests that are P.1 & 📈.

https://twitter.com/trvrb/status/1389248436071919628

2/ on sampling:
- we do more tests in some states. & N sequenced also based on pos rate. => can click on a state, or scroll on bar to see actual count.
- we do not select for SGTF or non-SGTF.
- we show fraction of P.1. to non-SGTF.

Weekly update from @my_helix SARS-CoV-2 surveillance program.

Check dashboards here: public.tableau.com/profile/helix6…

As reported by @CDCDirector, B.1.1.7 represents more than 50% of cases in US. And fraction still increasing.

See 7 states where Helix tests a lot. Details in 🧵 2/
Before going in details on B.1.1.7, here are some stats on vaccinations in US from @CDC. What an effort! I am very optimistic for the US.

Also important to note that all vaccines work very well against the B.1.1.7 variant. It is really incredible how well they work.

Update on @my_helix SARS-CoV-2 surveillance program

- Dashboards here: public.tableau.com/profile/helix6…

Some analysis on 7 states where we (Helix) have most positive tests 👇

Details in following tweets
(1 / n) 2/
How did I pick these 7 states?

- Each had >50 positives (Cq N gene <27) per day in recent Helix tests

- Alternative could be to use a threshold for % of positives in state coming from Helix
~5% of pos in these 7 states come from Helix

State Ns: covid.cdc.gov/covid-data-tra…

Update on @my_helix SARS-CoV-2 surveillance

- Seq data up to Feb 26. >15,000 sequences since Jan 1.
- Check dashboard for stats on tests & variants of concerns in your state. public.tableau.com/profile/helix6…

Re B.1.1.7, Michigan 📈 fast. MN too but we have less data.

More in🧵 2/
On Method:
- Numbers of positives, SGTF etc. are only reporting those with N gene Ct <27 (we use this threshold to select samples for sequencing).
- In Jan, we enriched sequences for SGTF to get grasp of % of SGTF that are B117. Now we don't enrich for SGTF anymore.

Update on @my_helix SARS-CoV-2 surveillance program

Check public.tableau.com/profile/helix6…

~50% of cases in Florida, Georgia & Texas are B.1.1.7.

in FL: Overall cases continue to 📉 but 52.8% of pos were SGTF (S-gene target failure) on March 7.

=> We will track closely evolution 2/
In Georgia

54% of positives were SGTF on March 7.

While still ~90% of these are likely B.1.1.7, interesting to note that B.1.525 also sequenced several times in GA and also leads to SGTF

(note: B.1.525 is not a variant of concern)

Update from @my_helix SARS-CoV-2 surveillance program

Check public.tableau.com/profile/helix6…

🧵 with FL, TX, GA, CA & MA + results on identification of B.1.351 & P.1 !!

FL: B.1.1.7 is ~40% of positives. 📈 about as predicted

Absolute numbers of B117 up despite overall 📉 in cases 2/
Texas

March 3: B117 were ~30% of positive tests.
26% with 5-day avg.

Absolute numbers of SGTF (a great proxy for B117 now) also going up despite big decrease in overall cases (as seen overall in US)

Note: this is only based on Helix data

Data drop from @my_helix SARS-CoV-2 surveillance

Check any state public.tableau.com/profile/helix6…

Or download from github.com/myhelix/helix-…

To see if large vaccination effort can keep overall cases from📈again -> important to monitor FL

See @ScottGottliebMD for perspective 2/
With overall number of cases 📉 quickly => important to look at % of positives that are B1.1.7
AND also the evolution of the absolute number of B1.1.7

Can do this based on Helix numbers,
or by multiplying % from Helix by overall number of cases reported by states and CDC