Skip to main content


The Latest on Coronavirus Mutations

For people looking for an accessible writeup on the coronavirus mutational landscape, I can recommend this Reuters article that came out today. It has a lot of good information in it, and a lot of very well-made graphics to show what’s going on. Past blog posts on this subject are here, here, here, here, and here.

And what’s going on, of course, is that the virus is mutating. It’s what viruses do. They don’t have a lot of overhead for lots of redundant error-checking machinery (although some have more of that than others), and honestly, total fidelity is not really an evolutionary advantage. A little sloppiness in copying the genetic material gives a more diverse population of viruses, with members that are more likely to be able to meet new threats or take advantage of new opportunities for infection. You would expect, over time, the viruses that can do a better job of that to be more represented. And remember, evolutionary time works different for viruses than it does for us. They turn around a new generation so quickly and in such huge numbers; they’re mashed down on fast-forward constantly.

So the mutational background is constant, but it’s important not to make the teleological error of picturing this all as being due to calculation, with the virus outwitting its adversaries. It can look like that, for sure, but it’s just millions of random chances spewing out everywhere – some work, some don’t, and what we see is the residue of some stuff that worked. You’ll see from the Reuters article that strains with a Gly in position 614 (the D614G one and its further offshoots) have become much more prominent. And those do seem to have a bit of an advantage in the binding behavior of the Spike protein – but if you look at some other situations around the world, you can see that other factors are at work. Singapore, for example, showed a lot of low-frequency mutations for a long time, apparently because these were showing up in foreign worker dormitory buildings which were then hit with vigorous quarantine measures. South Korea, for its part, had a lot more “V” family strains for a while due to a single superspreader event, which stood out against the country’s generally strong response. So there are a lot of extraneous factors and sheer accidents mixed into the data landscape.

The good news continues to be that none of the mutations studied so far in the general population seem to be able to evade the antibodies raised by the current vaccines. That doesn’t mean that it can’t happen – and as we start putting selection pressure on the virus by vaccinating people we’ll have to keep a close eye out for anything like that developing. But then we have to consider transmission. If an antibody-evading form of the virus also becomes harder to catch, well, it’s going to be less of a worry. But if we were to start doing a better job at not spreading the virus in general, that would be sort of nice, because that would reduce the chance that any nasty mutated forms get any kind of traction in general. If some sort of supervirus mutation occurs in a single patient who doesn’t then get close enough to other people for it to spread, then it’s a tree falling in a forest that doesn’t make much of a sound.

It’s all a race between several different factors. But here in the US we have so many people infected (and so much transmission going on) that frankly we’re making ourselves vulnerable to any more dangerous mutations that might crop up. In fact, if something like that were to emerge, the odds are better that it would do so here, from what I can see. We’re giving the virus every opportunity to reproduce and for the subsequent viral variations to then go out and try their luck infecting lots of other people. Vaccinating enough people quickly enough would interrupt these processes, and so would doing the sorts of public health measures that we’ve all been hearing about for months. But the first is going to depend on vaccine supplies, logistics, and public acceptance, and the second, well, look around you, si monumentum requiris.

It’s good news that no profoundly worse mutations have spread so far, and that might be a sign that they’re not so easy to come by. But then again, this virus has a relatively short acquaintance with human beings, so I think it would be prudent to assume that there are still a lot of things that we don’t know about its interactions with us. Let’s do everything we can to give it as few chances as possible.

104 comments on “The Latest on Coronavirus Mutations”

  1. FM says:

    Mr Lowe.

    I believe you have associated an incorrect link for the Reuters article.



    1. Marko says:

      If he’s like me , he had twenty tabs open on different stories across the web , and he copied and pasted the link from the wrong tab. I have to double-check myself on this all the time. My greatest fear is that one day I’m gonna post a link to my favorite porn site to a forum like this one.

      Hear the proper link to the Reuters piece :

      1. Not a rick roll says:

        That would’ve been the perfect moment for a Rick Roll

        1. Oudeis says:

          I’m pretty sure the perfect moment for a Rickroll passed in 2008. 🙂

          But okay, yeah, it would have been funny.

  2. Richard Collins says:

    The first link goes to a 2015 article entitled “Decline In Economic Returns From New Drugs Raises Questions About Sustaining Innovations”

  3. FoodScientist says:

    “It can look that that” paragragh 3 He’s ready for the weekend.

    1. Derek Lowe says:

      Pretty much! I had to write quickly and get to other tasks, and I’m only now going back and fixing things (and removing the digital spinach from between my teeth).

  4. myst_05 says:

    >> In fact, if something like that were to emerge, the odds are better that it would do so here, from what I can see

    Not that I disagree on the US having a lot of infections, but so does Mexico, Brazil, most of Africa, Russia, most of Europe… they’re just not as prominent in the US media and not all countries have a good testing infrastructure, but the US is not alone in having tons of cases/deaths.

    1. confused says:

      Yes, the US is relatively high on the scale but not as high as one might expect given the lack of federal response… several European nations have higher per-capita death rates, as do several Latin American ones.

      However, for mutations absolute numbers are likely relevant… the US doesn’t have the highest per-capita rate, but most of the harder-hit countries have much smaller populations (though Brazil is comparable).

  5. Bob Marlee says:

    This makes me wonder whether those developing the plans for group-based deployment are taking into account the prevalence of ultra-cold facilities, or making contigency plans for willingness to adopt. Health care workers obviously ought to get first dibs, and younger folk may be more likely to be asymptomatic spreaders, but if you are unable to reach some quickly because they are out of the way locations are reluctant to accept the vaccine, will they pump-up those who might otherwise be low down on the list but are near the cold chain and willing to take the vaccine? i.e; if optimal deployment ends up being difficult, will they settle for whatever deployment they can get?

    1. Tom Maguire says:

      The cold chain will be a challenge in many countries but in the US there are reasons for optimism. CVS and WalMart (among others) will be involved. WalMart has 5,000 pharmacies stocking up on freezers and dry ice and (they say) 90% of Americans live within ten miles of a WalMart.

      1. Bob Marlee says:

        Thanks, that is promising.

  6. theasdgamer says:

    I can’t believe that people are still pushing that asymptomatic superspreader bilge.

    If a cell is infected with SARS-COV-2, it’s gonna do nasty things to the cell. It’s not gentle like some viruses are.

    Pro Tip: Learn what “cytopathic” means.

    1. BB says:

      Can you please dumb this comment down for me a bit? I read the abstract to the linked article, but I’m still not sure I understand.

      1. CET says:

        It calls to mind a scene from a sci-fi horror movie (Slither if I recall correctly) in which an infected person collapses under the weight of the parasites inside them and then hundreds of thousands of the parasites tear through the person’s skin on their way out…

        Except that here it’s happening inside you, and I’d guess the number of exiting parasites is waaaay more than 10^5.

        Granted, I’m not sure that’s much worse than most other viruses. But I guess I know what my biochem students are reading next term. 🙂

    2. DiogenesNJ says:

      “Asymptomatic superspreader” is not “bilge”. But we have to include PRE-symptomatic.

      Since you’re dropping terms of art, you yourself could look up “serial interval”. It’s not a hard concept. When it’s possible to track the chain of infection, which South Korea for example did assiduously and prevented further confounding exposures by aggressive quarantines, then if the mean time to the appearance of symptoms is shorter than the mean incubation time, the infectious period must begin before symptoms do. Several studies have confirmed the time of maximum infectiousness is relatively short but starts 24 hours or more ahead of symptoms to a day or two after.

      Combine that with strong individual variation in the rate at which people emit possibly infections particles during speech. We also know that loud speech (noisy restaurants and bars, cheering at sports events and concerts, or singing) increases emission by an order of magnitude.

      So a “superspreading” event is a coincidence of unlucky timing with unlucky circumstances — a high emitter in an emission-enhancing environment at the presymptomatic period of maximum infectiousness. Both the genomics and the epidemiology support that.

      With regard to the main point of the article, which is viral evolution and mutation, consider this:

      While I normally ignore anything from Hahvahd, there’s a co-author from MIT which I regard as a pretty good school. So I guess I believe it. 🙂

      1. theasdgamer says:

        Let’s define terms:

        Asymptomatic–is infected and never displays symptoms

        Presymptomatic–is infected and will eventually display symptoms

        Neither of these has much impact on community spread. Presymptomatic MAY be a superspreader in a crowded indoor event. Not anywhere with excellent ventilation like a grocery store.

        Let’s consider a Chinese paper looking at Wuhan published in Nature Communications:

        They looked at 10 million cases and found 300 asymptomatic cases and no spreading by the asymptomatic cases.

        Asymptomatic superspreading is still a myth.

        1. a s says:

          Are grocery stores known for their excellent ventilation?

      2. Marko says:

        “With regard to the main point of the article, which is viral evolution and mutation, consider this:”

        That preprint you posted is pretty scary. I take some comfort in the fact that reinfections remain rare, but we should be on the lookout for geographic clusters of reinfections that might signal the emergence of an immune-evading mutation(s). I assume we’re making more of an attempt to sequence isolates from suspected reinfections than we were earlier in the outbreak, so hopefully we should be able to identify the troublesome mutations if they arise.

        The story could be somewhat different with the vaccines, of course, given their reliance on only spike/RBD immune response.

  7. DrivingAwayFast says:

    Mutations, you call that a mutation?
    is how you do a mutation.
    Please make it afraid of fire.

  8. Dale says:

    Derek – can you please do a post on the covid vaccines and pregnancy, eg, how pregnancy affects immune system of both mother/fetus and what that means for vaccine considerations? How those trials are run and how long the data takes to emerge?

    I think a lot of expecting couples are hunting for information as it is quite scarce. Any information is useful.

    1. Derek Lowe says:

      That’s the problem – information in this particular situation is indeed scarce! But I’ll see what I can round up in general. . .

      1. guineapig says:

        When I enrolled in the Pfizer trial, pregnancy and avoiding it was a big deal. They tested for it and we agreed to take steps to prevent it, including partner contraception. And yet they still managed to get a few…

      2. Dale says:

        Thanks. Even just “the basics” would be helpful. Anything to help people make informed decisions…

    2. ANON says:

      This is currently being studied. It was a topic that came up yesterday in the FDA’s committee meeting. Pfizer has already started a DART study. Data is expected in January.

      1. Irene says:

        Do DART studies cover breastfeeding safety? Currently I hear breastfeeding women in the UK are being advised not to take the vaccine. I suspect, given the safety profile of most vaccines, that it’s pure CYA and eventually that decision will be reversed, but of course I don’t know.

        1. ToxDox says:

          DART studies unfortunately do not cover nursing mothers. These studies are predominantly in animals to see if there’s potential harm to fetus or the mother. After a positive result, they will most likely move forward with pregnancy and lactation studies in humans.

  9. Valdis Andersons says:

    The good people at BioNTech and Pfizer have checked what some of the RBD mutants can do to vaccine induced antibodies (figure 1d):

    Thankfully, it looks like the vaccine induced antibodies are doing reasonably well against the various types.

    1. Michael says:

      Thanks — that’s compelling (and pretty cool).

  10. luysii says:

    As of 3 months ago —

    ““Researchers have catalogued more than 12,000 mutations in SARS-CoV-2 genomes. ”

    How many mutations are possible in the viral genome? Just 29,903 times 3, because at each position, the element normally there can change to only 3 others — the viral genome is made of RNA is a linear chain of only 29,003 nucleotides, and each nucleotide can be uracil (U), adenine (A) guanosine (G) or cytosine (C). That’s it. Proteins can have 20 different amino acids at each position.

    So 13% of all possible mutations have been found in the virus, out of only 90,000 completely sequenced genomes. There are now 28,000,000 cases out there, so it’s almost certain with 1,000 times more virus out there to sequence, that nearly all the other 44,000 or so possible mutations have already occurred somewhere in the world.

    How can this be good news? Because if any of them were truly horrible, we’d know about it. It would have taken over just the way the D614D mutation did.”

    Given what’s out there, is is almost certain that every single nucleotide mutation in the viral genome has occurred somewhere. Now all we have to worry about are double triple mutations and their possible epistatic effects.

    For more on this please see —

    1. Not a virologist says:

      I would have assumed that the risk is exactly from a virus having multiple mutations, rather than single point changes. Is increased infectivity in viruses typically from single base changes?

      1. luysii says:

        A single base change can do a lot of damage — see Derek’s post on the Rx of Sickle cell disease

        1. Not a virologist says:

          You’re saying “There exist cases where a single mutation is an issue”. I’m asking “In most cases where there’s an issue, was it a single mutation or multiple?” We’re talking about different things.

          1. luysii says:

            The best example of a mutation at a single location in SARS-CoV-1 having an effect is D614G as noted by Derek above. The following is from the link I supplied “Two SARS-CoV-2 viruses collected from anywhere in the world differ by an average of just 10 RNA letters out of 29,903,” If you call the 10 RNA letters mutations, then isolated single mutations do not occur in extant viruses. What must be done is disentangle the effects of a mutation at a given position, independently from mutations occurring elsewhere in the genome.

    2. Leo says:

      The math gets harder if you include insertions and deletions…

  11. Darn good post. Forced me to think about selection pressure vis-a-vis vaccination. And that the currency of the zero-sum game we are playing with the virus is *opportunity* to mutate. The more vaccinated, the less opportunity. No doubt epidemiologists could frame up some sophisticated formulae (much more easily than I could at least) to capture the effectiveness of the selection pressure of a vaccine vs the selection pressure of a therapeutic antiviral or for that matter a prophylactic antiviral.

    Anyway my concerns in a zoonotic bounce-back mutation that escapes vaccine coverage lies in regions where 1) animal husbandry is common, coupled with 2) a near-zero likelihood of effective Denmark-style culling of the animal population in question. An insufficiently vaccinated sub-Saharan Africa is the poster child for this persistent exposure vector to global public health.

    1. Crni says:

      Why would a zoonotic mutation be more dangerous than a human host mutation?

  12. Craken says:

    “In fact, if something like that were to emerge, the odds are better that it would do so here, from what I can see.”
    America is currently running at <20% of official global deaths. Therefore, the smart bet is that it's at least 4x as likely that a dangerous mutation will appear elsewhere. This bet is reinforced by the fact that America's official numbers are not running as far behind its excess death numbers as most of the world. But, perhaps a dangerous new mutation would be more likely to grow rapidly in America if it's deemed the largest single mutation market. The Reuters article shows the latest strain (GV) has grown much faster in more populous, quasi-united Europe than in America.
    I completely agree on the evolutionary reasoning for reducing spread and rapid vaccination.

    1. Marko says:

      ” Therefore, the smart bet is that it’s at least 4x as likely that a dangerous mutation will appear elsewhere….”

      Maybe , but if a dangerous mutation gets built on a backbone of previous mutations , it looks like the Anglo countries might be a likely place for that to occur :

      “The top 5 countries bearing all the high significant and majority of the moderate significant mutations are: USA, England, Wales, Australia and Scotland. ”

      That’s from July , so a little out-of-date , but it seems like some of the good mixing bowls for humans are also good mixing bowls for Cov-2

      1. M says:

        That sounds more like “Five of the countries for which we have the most data.”

        1. Marko says:

          It could be. You’d expect the countries that do relatively more sequencing to see more of the different variants. I’d like to know how the major countries stack up in regard to sequencing frequency. I know that Norfolk in The UK has just about everyone beat. They sequence a near-majority of their PCR-positives , I believe.

  13. ETHAN LOEB says:


    1) Has attenuation of SARS-CoV-2 pathogenicity been definitively ruled out yet?

    2) If attenuation of viral pathogenicity is partly contributing to the decreasing SARS-CoV-2 infection fatality rates being observed with time worldwide, could the universal masking, social distancing and lockdown measures taken to prevent the asymptomatic spread of attenuated viral strains in low-risk individuals have the unintended consequence of preventing or delaying further attenuation, not to mention causing the loss of life-years in the population from non-COVID19 mortality and morbidity as a direct or indirect result of these public health measures?

    In a recent recent episode of “doctor radio” which featured Christopher M. Petrilli, MD, who is an assistant professor at NYU and recently published a paper on the declining in-hospital mortality rates for SARS-CoV-2.

    Dr. Petrilli started the segment by stating that masks were partly responsible for declining in-hospital mortality rates. He was quickly reminded by the host that this was a hypothesis and not something that had been scientifically proven. At the end of the segment Dr. Petrilli stated that “the virus is not becoming less deadly“ without backing up that statement with any substantial scientific evidence.

    I’ve been trying to closely follow the literature regarding viral mutations and their relationship to pathogenicity and have come across many pre-print and peer reviewed articles that seem to indicate that the virus is mutating quickly and that different strains might be associated with different degrees of transmissibility and pathogenicity.

    In August, Dr. Anthony Fauci published a long paper on the SARS-CoV-2 pandemic in which he devoted one single paragraph to the concept of attenuation of pathogenicity and dismissed it as being unlikely at this early stage in the pandemic even though he admits it would be conventional wisdom for this to occur at some point due to natural viral evolution.

    I have created a Facebook page which contains the links to the studies that I’ve come across which support my hypothesis that if attenuation of viral pathogenicity is occurring, then public health measures aimed at preventing the transmission of attenuated strains of the virus in otherwise low-risk individuals may have serious unintended consequences.

  14. theasdgamer says:

    “this virus has a relatively short acquaintance with human beings”

    Oh, the “novel virus” myth. It’s just SARS with some changes. Otherwise we wouldn’t see a high percentage of people with T-cell immunity.

  15. Chris Phillips says:

    Here is a preprint by Ravindra K Gupta and colleagues, entitled “Recurrent emergence and transmission of a SARS-CoV-2 Spike deletion ΔH69/V70”, relating to the “new variant” of the virus that has been in the news in the UK:

    And here is an article with further comments from Gupta from the Guardian:

  16. Nate says:

    Can you elaborate more on cluster 5 specifically?
    I have a hard time believing that this hasn’t jumped given the number of mink farmers who were reported as sick with covid in Denmark and BC last month.

    1. JS says:

      As I read the updates from SSI in Denmark there have been no further detections of Cluster 5. However, many other mink related variants keep getting detected in Denmark. You can search on for details.

  17. John Moore says:

    Have any mutations increased transmissibility enough to explain some or all of the “third wave”?” Put another way, has R0 increased significantly?

    1. Chris Phillips says:

      It seems there is now a significantly more transmissible mutation spreading in south-east England. Strange not to see any comment here about it.

      1. Marko says:

        “Preliminary genomic characterisation of an emergent SARS-CoV-2 lineage in the UK defined by a novel set of spike mutations”

        “…Several aspects of this cluster are noteworthy for epidemiological and biological reasons and we report preliminary findings below. In summary: The B.1.1.7 lineage accounts for an increasing proportion of cases in parts of England. The number of B.1.1.7 cases, and the number of regions reporting B.1.1.7 infections, are growing. B.1.1.7 has an unusually large number of genetic changes, particularly in the spike protein. Three of these mutations have potential biological effects that have been described previously to varying extents:

        Mutation N501Y is one of six key contact residues within the receptor-binding domain (RBD) and has been identified as increasing binding affinity to human and murine ACE2.

        The spike deletion 69-70del has been described in the context of evasion to the human immune response but has also occurred a number of times in association with other RBD changes.

        Mutation P681H is immediately adjacent to the furin cleavage site, a known location of biological significance….”

        The consensus seems to be that it’s somewhat more transmissible by virtue of the way it’s spreading, and considering the currently-circulating comparator is D614G, which we already know is more transmissible than the earlier variants, it’s seems like we’re on the up escalator of transmissibility.

        1. Marko says:

          Though we’ve seen no solid evidence of increased severity of disease resulting from mutation, increased transmissibity is plenty bad enough. It means that each step forward in terms of herd immunity resulting from natural infection or incremental vaccination coverage may be matched or exceeded by these new transmissibilty mutations. Effectively , the R0 of each new higher–transmissibilty mutant strain may increase enough to outstrip our progress towards herd immunity. In the end, it may be that close to 100% of the currently-naive population will need to be either naturally-infected or vaccinated to protect those remaining that are still naive from infection and the possibility of severe disease..

          1. Chris Phillips says:

            I’d be interested to know if anyone can make head or tail of the statements made about transmissibility in the UK press conference yesterday, or if anyone is aware of any hard data from elsewhere.

            Apparently the new variant was said (1) to be up to 70% more transmissible and (2) to increase “the R number” by 0.4 or more (reported by some as 0.4 per cent!).

            And then there are also statements about the percentage of the new variant increasing from 28% in London and parts of the south east in “mid-November” to more than 60% in London in the week beginning 9 December. That looks more like an increase in the R number of 25% or so.

            Perhaps something else that made more sense was said and not reported by the media. I don’t know.

          2. Chris Phillips says:

            I managed to find the slide with the data these estimates were (presumably) based on. It’s a comparison between the percentages of the new variant in three English regions in the weeks commencing 18 Nov and 9 Dec. In London, it rose from 28% to 62%, in South East England, from 28% to 43% and in East England from 23% to 59%.

            Assuming a serial interval of 4.5 days, those numbers produce three estimates for how much larger the R number is for the new variant. The three estimates are 36%, 15% and 40%. Presumably 70% is the upper end of some kind of confidence interval. There must be a reason why the estimates differ so much – perhaps it’s partly because Kent, where the variant seems to have taken a hold first, has been under tougher restrictions than the rest of the South East for a while.

          3. DataWatcher says:

            R0 for measles ranges from 12 – 18, I believe. I’m guessing that even this new strain of COVID-19 isn’t approaching anywhere near that level, but how long did it take for enough people to take the measles vaccine to effectively eradicate measles?

          4. Marko says:

            I agree that we’re still a ways off from measles territory , but if you start with an R0 of 4.0, and layer on three sequential 50% transmissibilty increases , you’re at an R of 13.5 , and, voila! , we’re starting to think in a measles-like manner.

            On a global basis, with near-complete vaccine coverage expected to take years ( if it ever occurs ) there’s going to be plenty of time for this sort of viral evolution to occur, based on what we’ve seen so far.

          5. DataWatcher says:

            I was going to ask about the current R0 for COVID-19, as well. My understanding has been that it has remained between 2 – 3, for the most part. Has it been shown to be as high as 4.0, at least in some populations? And until now, has it shown signs of increasing dramatically?

          6. Marko says:

            Most people put the estimated R0 at the outset of the pandemic at ~2.5-3.0 , however, those estimates come from case data that was severely delayed and undercounted in the early days. My bet is that the initial R0, properly measured, would have been higher than 3. Still , for the purposes of my example , it doesn’t matter much. Just add a fourth ( fifth?, sixth? ) 50% transmissibilty increase to the calculation. You end up in the same mess, or worse.

          7. Marko says:

            I’m assuming that “transmissibility” means what it sounds like it means. If an infected person transmits to 3 people, on average, and transmissibilty increases by 50%, that person will now transmit to 4.5 people, on average. The effect of transmissibilty on R would thus be in the same proportions, by definition.

          8. Barry says:

            R0 by definition is the average number of new infections per case in the absence of public health measures. The R0 can vary from culture to culture, but is invariant over time.

        2. Marko says:

          ” The R0 can vary from culture to culture, but is invariant over time.”

          R0 defines the reproductive capacity of a given virus in a completely susceptible population, in the absence of mitigation measures. R0 is used to estimate the herd immunity threshold.

          If a virus evolves to transmit more efficiently, that virus would then have a higher RO in the context of a susceptible population with no mitigation, and thus a higher calculated herd immunity threshold. In other words, the herd immunity threshold can evolve as virus transmissibility evolves. The math is straightforward.

          1. Barry says:

            “R0 defines the spread of a given virus”. If it mutates, it’s a different virus, w/ a new R0

          2. Marko says:

            Haha. OK. I would call it a new “strain” of the same virus, whose new phenotype is increased transmissibility, but I’m happy to give you this one.

            We can call it SARS-CoV-2A (B,C,D,etc.)

          3. Marko says:

            “…Dr. Anthony S. Fauci, an adviser to both the Trump administration and the incoming Biden administration — has begun incrementally raising his herd-immunity estimate. In the pandemic’s early days, Dr. Fauci tended to cite the same 60 to 70 percent estimate that most experts did. About a month ago, he began saying “70, 75 percent” in television interviews. And last week, in an interview with CNBC News, he said “75, 80, 85 percent” and “75 to 80-plus percent.” ”

            “….Hard as it may be to hear, he said, he believes that it may take close to 90 percent immunity to bring the virus to a halt — almost as much as is needed to stop a measles outbreak. Asked about Dr. Fauci’s conclusions, prominent epidemiologists said that he might be proven right. The early range of 60 to 70 percent was almost undoubtedly too low, they said, and the virus is becoming more transmissible, so it will take greater herd immunity to stop it….”


  18. Marko says:

    Words that I predict will be proven wrong in fairly short order :

    As a blue-check “expert” you should have to openly eat your words when you get thing wrong, but this rarely happens. The mistaken “experts” simply close ranks, dig in, and continue to spread their errant takes. The guy above, like Balloux and others on the D614G variant, will be likely be mumbling “founder effect!” until his last breath.

  19. Chem. says:

    I saw a reference that a higher trans. mutant was in South Africa.

    1. Marko says:

      Yes, it’s a different variant than in the UK, though it does have the N501Y RBD mutation in common with the UK variant :

  20. Chris Phillips says:

    Unfortunately 70% is not an upper bound. The figure is discussed at 2:45 in this video:

    1. Chris Phillips says:

      However, the person presenting it says “it’s really too early to tell” and doesn’t seem to place any reliance on the 70% figure, only on the fact that the variant is growing very quickly. He notes that similar analysis for the A222V variant that came from Spain in the Summer had originally implied it had a significantly larger growth, but it turned out not to. He says that for A222V the model is not a good fit and may not be appropriate. As he says, the model is not a good fit for the new variant N501Y either. It looks a lot worse than for A222V to me. In the minutes of the NERVTAG committee advising the UK government, the figure is presented as “71% (95%CI: 67%-75%)”.
      If it’s really based on the work presented there, that gives a very misleading impression of the confident felt in the number.

      The other quantitative estimate in the NERVTAG minutes is:
      “Studies of correlation between R-values and detection of the variant: which suggest an absolute increase in the R-value of between 0.39 to 0.93.”

      Given that the distribution of the new variant within the UK is very inhomogeneous, and that different restrictions are in place in different parts of the country, I can imagine that it might not be safe to rely on that correlation.

      I don’t see how the 70% figure can be consistent with what has happened in London (one region under one set of restrictions). From late November to mid December the prevalence of the new variant has increased from 28% to 62%, but on the 70% figure I reckon it should have gone to 90%.

      1. Tony M says:

        I could be wrong, but I think the 71% increase is referring to a 71% increase in the growth rate ie (the expected number of cases directly generated by one case in a population). I think the infectious period for Covid-19 is 14 days. “From late November to mid December” implies a period of greater than 14 days, say 21 days or 1.5 infection periods. Therefore, expected percentage would equal 28% at start x 1.71 to the power of 1.5:
        = 28% x 1.71 ** 1.5
        = 28% x 2.24 = 62.7%


        1. Chris Phillips says:

          As I mentioned above, the time between the two percentages was indeed 21 days. But the 14 days you have used for the serial interval is far too long . I used 4.5 days. That may be slightly too low. This review suggests 5.19 or 5.4 days:

        2. Chris Phillips says:

          Also, you don’t want to use just the ratio 62% to 28%. You want to use that ratio for the new variant, divided by the equivalent ratio for the old one, namely 38% to 72%. That’s in order to get a ratio of the R values for new and old.

          1. Mariner says:

            It is worth considering that the conditions were different during November and December. There was an England-wide lockdown from 5th November to 2nd December when hospitality venues and non-essential shops were all closed.

            When this lockdown ended, London and most of the South East of England emerged into “Tier 2” of restrictions which allowed hospitality to reopen, with various rules to be followed – pubs and bars could reopen, providing the alcohol was served with a “substantial meal”. Cue ridiculous debates about what consituted a substantial meal with a Scotch Egg being considered enough for pubs without kitchens.

            Needless to say, with the relaxation of restrictions following a 4 week lockdown and the busy Christmas period approaching, I’m not surprised that there was a surge in infections. Whether or not the new variant was the cause of this or simply happened to be the strain best-positioned to take advantage of the end of more serious distancing remains to be seen.

            I am pretty relaxed at present about the likelihood of increased transmissability of this strain itself being the problem though obviously it may prove to be the case. I’ll wait to see further research before getting too worried and, in the interim, cutting down on mixing over Christmas was always the correct choice so will save lives regardless of just how much more dangerous this new strain might be.

          2. Chris Phillips says:

            That calculation based on the rise in the percentage of the new variant in London from 28% to 62% isn’t influenced by the overall growth in London over that period, though. That cancels out, because it’s a calculation of the ratio of the R values for the two versions of the virus.

            I agree that some of the other calculations may be influenced by different restrictions in force in areas where there is different prevalence of the new variant. But the restrictions are the same throughout London.

      2. Chris Phillips says:

        “I don’t see how the 70% figure can be consistent with what has happened in London (one region under one set of restrictions). From late November to mid December the prevalence of the new variant has increased from 28% to 62%, but on the 70% figure I reckon it should have gone to 90%.”

        And in fact thinking about the dates, it was probably only for 10-20% of the period in question that Kent, the county where the new variant was most prevalent, was under tougher restrictions than the rest of the South East. So on the basis of a 70% increase in R it should have got from 28% to the upper 80%s there too, and it got to only 43%, which would indicate only a 15% increase in R. It doesn’t seem to add up.

          1. Marko says:

            Another week or two of data from the affected areas should firm things up considerably, I’d think, unless incidence plunges so much from the lockdown that data effectively dries up.

          2. Marko says:

            Bottom line from the new report :

            “We have assessed this variant as having substantially increased transmissibility with high confidence.”

            I suppose all the countries shutting their borders with the UK must have a fair amount of confidence , as well.

            Certain blue-check “experts” : “But , but , wait , founder effect something something…”

          3. Marko says:

            When you’re in a hole, you’re supposed to stop digging, right? Apparently not, for some:


          4. DataWatcher says:

            Well, for what it’s worth, Moncef Slaoui also says that although there’s clear evidence that there”s more of the new strain in the population, this doesn’t necessarily mean that this strain is actually more transmissible, suggesting that the apparent increase may be an artifact of people having become more lax in following public health protocols, and/or the fact that scientists are now able to track variants of the virus more efficiently. Difficult to tell whether Slaoui is just being a cheerleader — but he did, after all, predict the 90%-plus efficacy of the first two vaccines, which at the time virtually everyone else thought was ridiculously optimistic.

            Meanwhile, the WHO’s Mike Ryan, while confirming that COVID mutates more slowly than seasonal influenza, also suggests that it’s unclear whether this new increase in spread is due to the mutation, or to laxity in “human behavior”. Maria Van Kerkhove, the head of the WHO’s emerging diseases and zoonosis unit, reports that the R0 for COVID (at least in the U.K.) has increased from only 1.1 to 1.5, which sounds really low compared to almost every other estimate I’ve heard.

          5. Marko says:

            “Maria Van Kerkhove, the head of the WHO’s emerging diseases and zoonosis unit, reports that the R0 for COVID (at least in the U.K.) has increased from only 1.1 to 1.5, which sounds really low compared to almost every other estimate I’ve heard.”

            First, don’t confuse R0 with “effective” R ( Re or Rt ). Re is a measure of what’s happening in a given community with various mitigation measures in place and with various levels of population immunity. Re tells you nothing about the herd immunity threshold.

            Secondly, an increase of Re of “only 1.1 to 1.5” is huge in its impact. At an Re of 1.1, political leaders would be making modest efforts to get the Re to 1.0 or below. At an Re of 1.5 , their hair would be on fire, and they’d be shutting down right and left , as is happening in London. (Well , maybe not Bolsonaro, but most with half a brain would be. )

          6. DataWatcher says:

            Okay, many thanks for the clarification. Still interesting, though, that Slaoui and Ryan, at the very least, seem a little skeptical about whether this new strain is quite as transmissible as some of the reports are indicating. I think you’re right in suggesting that we will simply have to wait a little while longer for the data from the infected areas to become clearer. And meanwhile, erring on the side of caution is far preferable to erring in the other direction.

          7. Chris Phillips says:

            Thanks. That’s interesting.

            The first calculation seems the most straightforward, and that produces a 47% increase in transmissibility. That is assuming a serial interval of 6.5 days. The result is really very sensitive to the value chosen for the serial interval. For 5.3 days the same calculation would give 37%; for 4.5 days it would give 31%.

            Estimates of the serial interval vary. I see that value of 6.5 days in the literature as a mean calculated for infector-infectee pairs ( But on a more fundamental level, it seems incorrect to use this simple mean time to characterise an exponential curve, because the time to infection is very variable: the paper giving 6.5 days as a mean also gives the standard deviation as 4.3 days.

            Will the exponential growth not be dominated by shorter-than-average infection times, by the nature of exponential growth? If so, the difference between the two strains would be over-estimated.

            So would the values of R0 itself, of course. Given that the value of R0 will vary a lot between different parts of the population, I’m already convinced by the argument that the simple homogeneous model will be significantly over-estimating the percentage required for herd immunity. Maybe the values of R0 are being systematically over-estimated as well?

          8. JS says:

            An increase in R is one explanation for the increase in the prevalence. Another is that the serial interval is shorter. Difficult to tell which, but important to know for estimating the vaccine coverage needed to achieve herd immunity.

        1. Marko says:

          “And meanwhile, erring on the side of caution is far preferable to erring in the other direction.”

          Exactly. That’s my primary issue with many of the so-called experts. Many of them seem to be “cautious” precisely in the wrong direction. We’ve seen it over and over, with masks, airborne transmission, D614G, etc., and it’s often the same cast of characters. When the evidence is compelling, but not 100% conclusive, their reflexive response is:” Move along, nothing to see here.”

          This UK mutation may well turn out to be a false alarm, but if it is, we’ll know shortly. In the meantime, treat it with the seriousness it deserves if it is , in fact, a real problem.

          1. RHB says:

            Re: “And meanwhile, erring on the side of caution is far preferable to erring in the other direction.”

            Problem with that ethos is where to stop . Had modern science been around at time, most of 20th century would’ve been stopped in tracks (and all centuries before that).

            Going back to early 1950s myself, that would be a a kinda shame. Could have been tame, or maybe even uglier than it turned out, who can tell.

            Whatever, as ever, times they are a changin’. Led it grow…


  21. Marko says:

    A great thread on antibody escape mutations in the common cold coronavirus, 229E:

  22. Martha Careful says:

    An ideal antigen should be selected for the development of a safe and effective COVID-19 vaccine. The S protein is the major antigen in most COVID-19 vaccine candidates under development as it contains the major neutralizing epitopes and is located on the surface of the viral particle. However, the full-length S protein of SARS-CoV also contains several immunodominant sites that can induce non-neutralizing antibodies, including those associated with ADE, or harmful immune responses78,79,83,84,119,120. For example, antibodies targeting the S597-603 epitope, which is located close to the carboxy terminus of the RBD of SARS-CoV S protein, markedly enhanced SARS-CoV infection of Vero E6 cells compared with antibodies from unimmunized macaques79

  23. Chris Phillips says:

    UK preprint entitled “Preliminary genomic characterisation of an emergent SARS-CoV-2 lineage in the UK defined by a novel set of spike mutations”:

    1. Frank Richards says:

      Thank you, Marko and Chris Phillips, for drawing attention to the scientific basis for the SARS 2 mutant strain identified over the last month in the South Eastern UK, and now forming basis of Threat Assessment Brief from ECDC (European Centre for Disease Prevention and Control):

      For the record, Figures 1 and 2, included to support arguments in the Assessment Brief, are not readily directly comparable and contain no error bars. If suitable graphs suitably superimposed would error bars overlap? Just asking.

      Also for the record, work initially disclosed on, a discussion forum site with 37 users, which presumably includes the 10 authors of the initial disclosure, of whom the lead author is site “Admin”:

      Further for the record, no sign of any independent peer review. Common feature of some relevant scientific papers throughout SARS 2 epidemic, “The Science” and the “Scientific evidence” oft cited here in UK by Government as “driving policy.” Just saying.

      Prime example being Imperial College paper published 16 March 2020, which over a single weekend changed UK government policy by predicting 510K UK deaths (official number to date sadly ~60K), yet used word “assume” (or derivative thereof) around 40 times and assumed no populace prior immunity (arguably contradicted by subsequently emerging T cell hypothesis). Another critical paper not subject to independent peer review:

      Sense of déjà vu, as over weekend latest mutant strain now spreading panic in UK and European governments. This morning’s BBC headline reads, “Covid-19: Boris Johnson to chair emergency meeting amid travel bans.”

      France now withholding permission for freight to cross from Dover for 48 hours, which in turn might mean continental lorry drivers won’t cross to the UK for fear of getting stuck on wrong side of channel. “Concerns raised that border restrictions could disrupt UK food supplies.” Implication UK to be cut off at the Channel. Going out anyway, so will recce shortly for signs of panic buying…

      All plausibly entwined with UK Prime Minister’s BREXIT homework, set to himself four and a half years ago when leading referendum campaign to take back control, and still not handed in complete at 11.59th hour of BREXIT negotiations (btw, ditto Biology virus homework left undone in run up to February half term, but that’s another story…).

      Weekend headlines conveniently dominated by mutant strain. BREXIT negotiations nowhere to be heard of in UK media. Nemesis plausibly looming, or will shadowy scriptwriters, parliamentary sheep and cabinet stooges somehow combine with Legendary Leader to manufacture latest Finest Hour when plucky Nation in true Dad’s Army style somehow keeps European Cataclysm at bay..?

      Or as the errant schoolboy Billy Bunter might have explained to Mr Quelch in long ago boarding school yarns set in the Lower Fourth Form at Greyfriars School, class known in parlance of era as “The Remove…”

      …Sir, Sir, a breath of quantum wind in my room blew away my homework out of the open window Sir, ‘cos it was an unusually hot enough day in the middle of winter to have the window open Sir, then the school dog must have been running around outside and eaten my homework Sir, it’s true Sir, it must have really happened Sir, honest it must have, mustn’t it?

    2. Marko says:

      A corresponding paper describing the South African lineage :

      Emergence and rapid spread of a new severe acute respiratory syndrome-related coronavirus 2 (SARS-CoV-2) lineage with multiple spike mutations in South Africa

      1. Marko says:

        A nice illustration of the locations of the mutations across the spike protein :

      2. Marko says:

        Table comparing the two variants with the 501Y Spike mutations in South Africa (501Y.V2) and the U.K. (B.1.1.7) :

  24. Chris Phillips says:

    There seem to be few hopeful aspects of this situation, but Lucy van Dorp of UCL comments (among other not-so-good signs):
    “B.1.1.7 also harbours a truncated ORF8 gene, with deletions in this region previously associated with decreased disease severity.”

    There is a link to a paper which actually shows a very dramatic decrease in severity:
    “Development of hypoxia requiring supplemental oxygen was less frequent in the ∆382 variant group (0 [0%] of 29 patients) than in the wild-type only group (26 [28%] of 92; absolute difference 28% [95% CI 14–28]). After adjusting for age and presence of comorbidities, infection with the ∆382 variant only was associated with lower odds of developing hypoxia requiring supplemental oxygen (adjusted odds ratio 0·07 [95% CI 0·00–0·48]) compared with infection with wild-type virus only.”

  25. Chris Phillips says:

    The BBC reports comments by members of the UK NERVTAG committee about indications that the new variant may be infecting children more easily, and that (according to Neil Ferguson) although it is unproven “”If it were true, then this might explain a significant proportion, maybe even the majority, of the transmission increase seen”:

    This would be consistent with a recent rise in infection rates among children in the UK infection survey.

    1. RHB says:

      C’mon, don’t encourage GOV.UK, Chief Medical Officer, Chief Scientific Adviser and SAGES to do more of their worst. More than enough nursery, school, college and university days lost already since GOV.UK declared war on MARS (Marked Acute Respiratory Syndrome) on 16th March 2020 (a day that will live on in infamy…).

      Back of envelope “Remote Learning Days,” since day of infamy for UK’s estimated 9 million 4-18 year olds and 1.2 million 19-22 year olds in higher education, comes out at 648 million school RMDs and 132 million higher education RMDs.

      Some total RMDs that. With estimated one or two hundred million parent HWDs drawn into the fray, where Home Working feasible, otherwise taken as Days Off…

      …Otherwise known as DOs, which can of course, after weeks on end cooped up in same household to SAVE LIVES, STOP the SPREAD and PROTECT the NHS, descend into an overprescription of DONTs.

      Do wonder if many commenters In the Pipeline have kids themselves (I do, two now grown up), or is it mostly theorists commenting in vacuo, like of which that Einstein quote deplores (or was it one of Oppenheimer’s?)?

      Latest Times headline today here in UK: Mutant virus is “everywhere.” Chief Scientific Adviser saying, “toughest restrictions must be extended in area.”

      Public Health officials instructing people travelling elsewhere from Tier 4 areas in south east England over Christmas to, “Self-isolate for 10 days.” BBC reporting that, “Boris Johnson acknowledging for first time schools might not re-open in January.”

      Meanwhile, on inside pages Times headline reporting, “Foreign experts scoff at British response to mutant virus,” as, “researchers in other countries accused Britain of over-reacting.”

      Scientists from other countries duly chip in regarding threat posed by mutant strain, “Not clearly established at this stage… Taken aback by drastic measures… Reserving judgement…”

      So there we have it. Getting distinct impression UK general public no longer listening. I am. For entertainment value. And For The Record.

      1. Chris Phillips says:

        I think most people come here with a scientific perspective, not for a political one. While science can inform the political discussions, the converse is much less true. It usually just generates crap for people to wade through.

        1. RHB says:

          I too come here for the science. For the record, I’m a scientist through and through – medicinal chemist at major UK healthcare R&D site for 32 years, first 20+ years making compounds at bench using my own hands. 40+ Published scientific papers (25 as lead author), named inventor on 40+ granted patents, 10+ clinical candidates, associated with discovery of an anticancer drug now considered best available for subset of lung cancers.

          But distanced in retirement 6+ years from real life R&D, I can no longer see science as an island entire of itself, but as a piece of the continent, a part of the maine.

          So for what it’s worth, coronavirus mutation science has had immediate high impact factor on country of 67 million and on the country’s international relations. Rightly or wrongly, I don’t know for sure, but know which way I’m leaning.

          Two days ago I posted a previous comment, that began with a couple of general scientific observations on the multicentre coronavirus mutation paper posted on the discussion forum that you’d highlighted – absence of error bars and absence of independent peer review (hence Times quote included in preceding comment about “researchers in other countries”)

          On screen message said comment would be subject to moderation and comment may by now have been rejected. Fair enough (comment ended pretty politically…). FYI, comment had been distilled from a longer item that looked more widely at peer review of critical epidemiology papers relating to the SARScov2/COVID-19/SARS 2/MARS epidemic. Decided not to post that in the end as too long, but compiling helped to clarify thoughts.

          Thank you for posting links to publications and for data you’d pulled out from other online material. Very helpful.

  26. JS says:

    Analysis of N439K variant:
    In Danish, but translation programs should work.

    No indication of higher infectivity or morbidity, but some concern about reduced sensitivity to existing neutralizing antibodies.

  27. Marko says:

    If this is accurate, it’s another damning indictment of the US pandemic response :

    “Since Dec. 1st, 31 SARS-CoV-2 genomes have been collected, sequenced, and deposited into public database from the entire United States. England has ~2,000 genomes over the same time period. We really don’t know if the variant is present in the USA.”

    On a population-adjusted basis, the UK appears to be sequencing at a rate ~300x that of the US. No wonder we aren’t seeing these mutations – we’re not looking for them!

    Little Iceland, with one company (DeCode) doing the bulk of the work, somehow manages to sequence virtually EVERY positive swab. Of course that’s a bit easier to do when you’ve had an intelligent response to the pandemic, and kept your case rate ( and death rate ) low.

    Still, the US has more pandemic “experts” who are social media sensations than any other country, by far (including our self-proclaimed “Earth’s Virology Professor”).

    So there’s that….

    1. JS says:

      One lab in Denmark sequences 5000 samples per week. The head of the lab says that they sequence as many in a day as Germany has done in the entire epidemic. Perhaps that is why Denmark has seen ~9 cases and Germany none?

      1. Marko says:

        No doubt.

        I suspect the reason Fauci advised against shutting down UK travel into the US is that he figured that the horses have already bolted from the barn. Our scientists can’t see all those horses though, because they’re wearing sequencing “blinders”.

  28. Billy says:

    Wow what luck.
    A virus that waited to mutate the instant the vaxxes were approved. And how lolite for each mutation to be located within specific country borders.
    Couldnt be that thus virus story is planned to laat 5 years, and the population will be constantly getting jabbed for the previous virus strain?

    The Invisible Rainbow pdf.
    The Poisoned Needle pdf.
    Pasteur vs Bechamp
    Germ vs terrain theory
    Somatid pleomorphs

    You fear of viruses will disappar once you learn the science facts instead of this germ theology which had to ignore the Koch postulates (and never be isolated) to be sold to you!

Leave a Reply

Your email address will not be published. Required fields are marked *

Time limit is exhausted. Please reload CAPTCHA.

This site uses Akismet to reduce spam. Learn how your comment data is processed.