Hacker News Reader

Comment by Tycho | Hacker News Reader

Wouldn't there be an element of cherry picking to this whole variant story? Like, don't viruses mutate constantly, we just don't analyze them in a disciplined manner most of the time, so if you start doing that you're bound to find something like this, but so would anyone else conducting a similar analysis anywhere else and at any other point in the pandemic?

View on HN

Yes variants occur all the time but they don't tend to become dominant. This variant went from 30% of all infections in November to 60% three weeks later. In order for a given variant to supplant all the others and become dominant it must have some characteristics that make it more transmissible. That is why they are particularly worried about this variant.

Could it still be that this variant is as infectious as others, but just happens to be the dominant one in the regions where the virus circulates the most, for reasons that have nothing to do with the genetics of the virus?

London and the South East are some of the densest populated areas in the UK, and one could expect exponential growth of infections there while other less populated areas could manage to keep their Rt around or bellow zero. If this strain was more prevalent in these regions, you would also see it taking a larger share of the infections nationwide.

However, the precautionary principle has been the keystone of good handling in the pandemic, so they are right to apply precautionary measures before it's too late. We will learn more about this strain in the next few weeks.

I don't know enough about epidemiology but I would imagine they have done modelling to determine the likely transmissibility or R number of this new variant to be 70% higher. The probability that this new variant has come to dominate by pure chance must be small.

You are right that London is a densely populated area prone to easy spread for the virus but the same must be true for all variants. This variant started its existence as a single strand of viral DNA and has managed to spread far enough to become the dominant strand against competition from many other well established variants.

This variant was not dominant in London some weeks ago, and now is.

Look at the prevalence graph in the article, you can see that the new variant has been gradually taking over the turf from the other ones. This cannot be explained by just the founder effect unlike most other cases, since the prevalence was high to start with. It cannot be explained away by a single super-spreader event, since a single event will just cause a single step-change. This has been a continuous process.

It could be random chance or a selective advantage, but then it comes down to just a modeling exercise. How likely is it that this could happen by chance? And it appears quite unlikely: instead the best way to explain the data is a significantly increased transmission.

Is this also taking out potential confounders out of the equation? I believe the currently available data (as opposed to the SAGE minutes, which has the conclusions) is not sufficient to rule that out.

What confounders would you suggest? Elsewhere in the thread you've been suggesting it's a founder effect. It should be plainly obvious why it's not that, nor something you could attribute to a single super-spreader event.

Reading the latest document out, I can't rule out sampling bias, and the methodology used to gather data is also vulnerable to bias (get samples which are PCR-negative for the S gene, but positive to other genes, which is, by their own admission, a poor proxy).

The confidence intervals shown by PHE on potential increased transmissibility are also very wide (not the ones from the NERVTAG minutes, but the new analyses by PHE).

It needs larger sampling (already doing so, I'm sure) and some biological evidence.

How would sampling error lead to the relative prevalence growing from 0% to 60% in a smooth exponential curve over 1-2 months?

My understanding on data gathering is that there have been two data sources: sequencing a 10% sample of the positive results, and using the fortuitous point about the three-target PCR tests showing one of the targets as negative for all such tests. Having two data sources is useful since the results from the sequencing are delayed by weeks.

But up to the point where they have both sets of data, the relative prevalence lines up very neatly. In particular, it cannot be that these results are coming from some other variant with the same 69-70 deletion.

(I don't think it's fair to suggest they implied it was a poor proxy in general. They said it was a poorer proxy the further back in time you go.)

Re: confidence intervals, the data they have from the relative prevalence of the new variant has pretty tight confidence intervals (95% CI: 1.34-1.59 R). That makes sense, because the modeling for that is really simple.

The confidence interval for trying to correlate prevalence of the variant vs growth rates is indeed quite wide. But it makes sense, because that's noisy data.

Yes, more data and more evidence will always be great. How many weeks are you willing to wait for it, before putting in new measures? How will that delay affect the epidemic curve if the findings so far are correct?

> Could it still be that this variant is as infectious as others, but just happens to be the dominant one in the regions where the virus circulates the most, for reasons that have nothing to do with the genetics of the virus?

Yes, but one assumes that confounding factors like that are fairly easy to isolate. At some point, especially as the growth continues, statistics pretty much rules out other explanations. General consensus is that this is almost certainly more infectious. But sure, there's always more science to do.

That can happen when general incidence is very low, in a small region, or for a short time. But this is not any more the case here, and having this happening by chance is very unlikely.

Sure, there's constant mutation and most of it is phenotypically meaningless. What caused the concern is that this particular set of mutations seem to be associated with a part of the country where the measured dose-response of lockdown measures suddenly went way down. In other words, the effectiveness of a particular set of lockdown measures seemed to suddenly go down in Kent and London but not other parts of the country including parts of the country that had previously shown similar responses to lockdown measures. Either everyone in Kent is suddenly breaking the rules more than people elsewhere, this is unfortunate random variation driven by a few super spreading events (totally possible), or this is the first phenotypically distinct variant (as opposed to irrelevant sequence mutations) which has emerged.

Let's hope it's nothing but I'm glad that measures have been taken now.

AFAIK, this strand is different because of the number of mutations. See figure 2 here https://virological.org/t/preliminary-genomic-characterisati...

This is a really good report on the new strain, and I urge everyone to read it.

Also, the UK seems to do way more sequencing of the virus than any other country, so one would expect such cherry picking to actually happen in the UK.

According to the Danish Serum Institute, the UK is sequencing about 10% of the positive test results. The UK is considered a leader in this field in Europe including the British Isles.

For comparison, the Danish Serum Institute has a sequencing capacity about 5000 positive tests a week, a rate of around 25% of the positives at the current level.

Source, in Danish, from the Serum Institute:

https://www.ssi.dk/aktuelt/nyheder/2020/ny-covid-virusstamme...

New Zealand is achieving over 80% sequencing and more recently has gone over 90%. It's possibly more important when you have less cases so that you can find connections and react to them. It's also a lot easy to do when you have very few cases (<10 versus 30k+).

https://www.rnz.co.nz/national/programmes/checkpoint/audio/2...

Also the UK detected this new strain right next to the Dover Crossing - Isn't it pretty likely it came from mainland Europe and was detected at our port rather than we coincidentally had a mutation right on the border?

Looks like it's probably the opposite.

> The variant can be found across the UK, except Northern Ireland, but it is heavily concentrated in London, the South East and eastern England. Cases elsewhere in the country do not seem to have taken off.

> Data from Nextstrain, which has been monitoring the genetic codes of the viral samples around the world, suggest cases in Denmark and Australia have come from the UK. The Netherlands has also reported cases.

Wales was reporting this variant as seen in all regions on Sunday, hence the lock down.

It is more likely that this is noted first in the UK. This means that there is some likelihood that it already exists in other countries, but was not notices yet.

However, this does not explains away that it appears to have much higher transmission.

As far as I understand, that is what happened with previous concerns about mutated strains. In the end they turned out to be no different than the wildtype, and greate distribution was simply because those strains were present in populations that caused more spreading.

There is not enough data right now to be sure whether this new strain is more infectious or not, but from what I read there are a few more reasons to be concerned this time compared to the previous times when mutated strains were reported.

My layman's interpretation is that variants are kind of a stepping stone to a strain. Viruses do mutate constantly, but I would guess that many (maybe most) of those mutations are effectively a no-op. It's a mutation in a dormant region, so it does nothing. Of the mutations that occur in an active region, some will be bad, and those mutations die out.

So I would think we're really only concerned with mutations in active regions that impact viability. We don't really care about mutations that decrease viability, since they'll generally be overcome by more viable versions of the virus naturally.

The question with this particular mutation is whether it's a variant or a strain. I.e. is this new variant more infectious? Which would make it a strain, not a variant.

Not really. Thousands of variations have ben sequenced, the notable thing here is the (potentially) faster spread.

Yeah, i read somewhere 70%.

Whilst the consensus seems to be that this variant is spreading faster than other variants - is there any information on exactly how it does so?

Not really. But this variant does have novel mutations in sections coding for spike proteins. Whether and how those are functional improvements for the virus are tbd.

See https://www.biorxiv.org/content/10.1101/2020.12.16.423118v1

Increase in the number of spikes per virus.

Just in time for Christmas too.

Reply on news.ycombinator.com