It's incredibly thrilling to have this paper out at last! πŸŽ‰

"Spread of a #SARSCoV2 variant through Europe in the summer of 2020"

(aka The Tale of Europe, Summer Travel, & EU1)

But why should we care about summer 2020 now? Let's take a look...

nature.com/articles/s4158…

1/24
Last Sept, while looking for #SARSCoV2 sequences that could help us understand transmission across #Switzerland, I noticed a cluster that was present not just across Switzerland, but also the UK & Spain. This is the cluster that eventually came to be known as 20E (EU1).

2/24
EU1 had a mutation at position 222 in spike - this caught my eye.

From mid-summer 2020, EU1 (orange) expanded across Europe - becoming the most prevalent variant in most of Western Europe, & accounting for >30% of sequences in Europe by the end of 2020.

3/24
We can see how EU1 rose over summer 2020 by looking at % of sequences each week that fell into EU1.

From initially expanding in Spain πŸ‡ͺπŸ‡Έ, it quickly started being detected in other countries soon after EU borders opened in mid-June. βœˆοΈπŸ”›

4/24
We see a general pattern of a rise, then a plateau in proportion, & a wide variation in how much EU1 dominated in different countries: in Spain, Ireland, & UK it was >60% sequences - in Norway & France it was <=20% by end of 2020, with other countries in between.

5/24
We were able to confirm Spain as the origin of expansion through phylogenetics - a collapsed phylogeny of seqs up to 30 Sept shows countries with shared genotypes as pie charts.

The earliest diversity is found in Spain (red) & shared with many other countries.

6/24
My initial concern was that EU1 was what we were all fearing in 2020: a more transmissible variant. 😱

But laboratory work showed no difference in antibody binding (pic 1) or pseudoviral titers (pic 2).

So how did it spread so successfully?πŸ€”

7/24
The summer of 2020 followed on from one of the hardest springs in memory.
On heels of lockdowns, economic hurt, hospitalizations & death, it was defined by hopeful restriction loosening & a longing to regain normality.

Could travel have played a role in EU1's success? πŸ§³πŸ›«

8/24
We were able to look closely at #SARSCoV2 cases across Europe over the summer (pic 1), including cases by Spanish province.

In addition, we looked at travel (pic 2). We usedπŸ“±roaming data from across Europe to see when & where in Spain visitors went.

9/24
By combining the case numbers per province over time😷, with the location data πŸ—ΊοΈ, plus info on their home country epidemic, we knew (generally) where people came from, where they went in Spain, & when.

We could use this in a simple model to predict how EU1 spread:πŸ”€

10/24
While this matches the dynamics fairly well, our model was underestimating the proportion of EU1 cases.

In some countries we did well (France), in others off by 2-4x (pic 1). But in others we were off between 7-11x (pic 2)!

What wasn't our model capturing? πŸ€”

11/24
When we compared our estimated travel-related cases with those reported by the German & Swiss govts (only German shown), we confirmed that our model was underestimating the likelihood of bringing EU1 home.

Our model really was simple: there's plenty it doesn't capture...

12/24
For example: people don't go to the same places in Spain on holiday! French visitors stay north πŸŒ„, whereas UK holidaymakers head south & to islands πŸ–οΈ (pic 1).

Countries have similar & differing travel patterns (pic 2).

13/24
See the paper for full discussion, but what our model's underestimation suggests is that travel related activities are associated with more cases than expected:

Demographics, behaviour, & variation in incidence (city vs province-level) likely play a role.
πŸšΆπŸ»β€β™€οΈπŸ½οΈπŸŠπŸ»β€β™€οΈπŸ•ΊπŸ»πŸ’ƒπŸ»πŸ‘¨β€πŸ‘§

14/24
But this was last summer... & EU1 isn't even more transmissible. So why should we care about it now?! 🀨

As we head once again into summer & the balance of travel & transmission, while hearing about new variants -
I think EU1 can still tell us some very important things!

15/24
EU1 underscores the potential impact of travel.

Last summer in Europe travel persisted as cases rose in Spain. There was essentially no testing, & quarantine may not have worked as well as we wished. Test&Trace didn't stop transmission chains.

16/24

ft.com/content/d214ef…
Differing travel volumes, locations & behaviours likely played a role.

Similarly, when we look at the spread of the Delta variant now in the UK vs other European countries, closer ties to India likely meant the UK has had many more introductions.

17/24

nriol.com/indiandiaspora…
EU1 also highlights that bans to just one country probably aren't enough on their own:
By end-Sept 2020, there was more EU1 in Europe outside of Spain than inside of Spain.
It wasn't about Spain anymore: other Europe countries were transmitting after failing to control it.

18/24
But perhaps most importantly of all, EU1 shows us that not all rising frequencies are due to viral change.

EU1 rose in frequency when cases were rising, & had a spike mutation - an alarming combination. But it was not a variant of concern - rather a travel opportunist.

19/24
As we identify new variants, the critical question is to what degree viral changes are influencing transmission & cases, & how much might be changes in restrictions, behaviour, introductions & more.

EU1 reminds us how important it is not to underestimate the latter!

20/24
EU1:
- Underscores the potential impact of travel particularly when incidence differs between countries πŸ›«πŸ§³
- Highlights what didn't work last summer πŸ˜·πŸ–οΈ
- Reminds us that not all rising frequencies are due to viral change! πŸ“ˆπŸ¦ 

21/24
This work thanks to:πŸ‘πŸ»
@MoiraZuber @SarahNadeau19 Tim Vaughan @khdcrawford @C_Althaus Martina Reichmuth, John Bowen @coronalexington Davide Corti @VirBiotech @jbloom_lab @veeslerlab David Mateo @kidodynamics Alberto Hernando @icoes @fgonzalef @TanjaStadler_CH @richardneher

22/24
I'm absolutely honoured to have worked with so many amazing people on the EU1 story.

They not only made it possible through collaboration & data sharing - they made it better & made it a true pleasure to work on. πŸ™πŸ»

In this pandemic, the only way forward is together. πŸ‘©πŸ»β€πŸ€β€πŸ‘¨πŸΎ

23/24
A short informal 'press release' covering the new paper can be found here:

docs.google.com/document/d/1b0…

24/24

β€’ β€’ β€’

Missing some Tweet in this thread? You can try to force a refresh
γ€€

Keep Current with Dr Emma Hodcroft

Dr Emma Hodcroft Profile picture

Stay in touch and get notified when new unrolls are available from this author!

Read all threads

This Thread may be Removed Anytime!

PDF

Twitter may remove this content at anytime! Save it as PDF for later use!

Try unrolling a thread yourself!

how to unroll video
  1. Follow @ThreadReaderApp to mention us!

  2. From a Twitter thread mention us with a keyword "unroll"
@threadreaderapp unroll

Practice here first or read more on our help page!

More from @firefoxx66

3 Jun
I am incredibly grateful to have received the 1st dose of the #COVID19 vaccineπŸ’‰today!

A year ago, I never imagined we'd have multiple vaccines available & being distributed now.
At the same time, I'm saddened the privilege I've been granted is still unavailable to so many.

1/4
This is often particularly the case in countries where continuing high case numbers & outbreaks mean lives are still being lost & people being hospitalized. We must continue to ask countries to do more to realise equitable & fair vaccination access.

2/4

nature.com/articles/s4156…
This is esp true for Western countries which are starting to reach high vaccine levels.

I call on wealthy countries to propose a plan for how they are going to contribute to better & more equal vaccine distribution.πŸ“’πŸ’‰

We aren't safe until all of us are safe. πŸ€²πŸ»πŸ§‘πŸΏβ€πŸ€β€πŸ§‘πŸ»πŸ§‘πŸΌβ€πŸ€β€πŸ§‘πŸ»πŸ§‘πŸΎβ€πŸ€β€πŸ§‘πŸ»

3/4
Read 4 tweets
14 May
Transparency & open communication is *so key* in this pandemic.

At @nextstrain we don't claim to be perfect - but you know who we are, how we work, & we try incredibly hard to have open conversations with the community: building trust & better research.

#openscience #opensource
It can be easy to say things like "we are too busy to tweet" or "we don't have time to talk to everyone." I πŸ’― get it! Not every one of our emails is answered, not every discussion post responded to. But if you want to truly be part of a community: you have to open up to it. πŸ‘πŸ»
It's painful sometimes: it means being open to change, being transparent about how everything works & how decisions are made -- it means openly crediting & celebrating others' work and contributions. πŸŽ‰ But all of that is vital in building a stronger scientific community. ⛓️
Read 4 tweets
2 May
πŸ—“οΈCoVariants.org is updatedπŸ—“οΈ, with some cool new additions:
- B.1.617.1/2 are added as 20A/S:154K & 20A/S:478K πŸŽ‰
- Beautiful new name table πŸ“‘πŸΎ
- Mutation list displayed in full as a "side-sausage"🌭

Let's take a tour... 😁

1/7
I know this took a little time (thank you for your patience! πŸ™πŸ»), but B.1.617 is now in CoVariants as the two sublineages 617.1 & 617.2, called respectively:
- 20A/S:154K (has 484Q)
- 20A/S:478K (no 484Q, has 478K)

2/7
We can see the two lineages best in the India graph of the 'Per Country' page - with S:154K in brighter green, & S:478K in darker green. (Note sequencing may not be representative)

They also show up in low numbers in some other countries.

3/7

CoVariants.org/per-country
Read 8 tweets
28 Apr
From the 'USA state' view (new) at CoVariants.org, I think the dynamics of 20C/S:484K (B.1.526) are intriguing.

Most variants that were circulating at end-2020 have declined after 501Y.V1 (B.1.1.7) (red) expanded - very noticeable with 20C/S:452R (purple) in CA:

1/7
But in states with 20C/S:484K (pink), this has seemed to grow alongside 501Y.V1 (B.1.1.7) (red) - sometimes growing, other times holding steady, or at least not declining significantly (may yet happen).

2/7
Some preprints have estimated the transmission of 20C/S:484K to be similar to 501Y.V1 (B.1.1.7).
20C/S:484K carries the 'Eek' 484K mutation in spike, which may be why neutralization (immune recognition) is lower.

medrxiv.org/content/10.110…
biorxiv.org/content/10.110…

3/7
Read 7 tweets
23 Apr
IT πŸ’» pulled out all the stops & the oven (server) is back online, so CoVariants is back in business!! πŸ› οΈπŸŽ‰

CoVariants.org is now updated - latest focal builds are in the oven πŸ₯–πŸ₯ -- & a very cool new breakdown feature has been added! (Read on!)

1/5
Overall variant frequencies per country sometimes don't reflect the whole picture. As a first step to hoping to support more regional views, you can now select a "USA" view on CoVariants' "Per Country" page & view by state!:

2/5

master.covariants.org/per-country
As always, be cautious when interpreting:
- Last data points, when data may still be coming in
- Data points/states with low numbers of samples - may not be representative

3/5
Read 5 tweets
1 Apr
β€œI am precariously employed; I don’t have a long-term job. I feel a lot of pressure that this is my opportunity and I cannot waste that" says Hodcroft.

Fantastic article by @meredithwadman on how scientists are coping (or not) in this pandemic marathon.

sciencemag.org/careers/2021/0…
I know PIs whose valuable pandemic staff are quitting because *they just can't take it anymore.* I know academics who are counselling post-docs while juggling next-day demands from governments - & of course balancing a family & sometimes homeschooling kids.
It's not sustainable.
You may have noticed I've been absent from Twitter a lot the past few weeks. It's the same phenomenon. I'm lucky: I have no staff or dependents. I'm gainfully employed. But I've spent more than a year feeling guilty *every waking moment* that _I'm not doing enough_.
Read 5 tweets

Did Thread Reader help you today?

Support us! We are indie developers!


This site is made by just two indie developers on a laptop doing marketing, support and development! Read more about the story.

Become a Premium Member ($3/month or $30/year) and get exclusive features!

Become Premium

Too expensive? Make a small donation by buying us coffee ($5) or help with server cost ($10)

Donate via Paypal Become our Patreon

Thank you for your support!

Follow Us on Twitter!

:(