There is one, and only one strain of SARS-CoV-2

coronavirusThe moment a preprint emerges describing a new patient isolate of SARS-CoV-2, with a change in the genome sequence, the world seems to explode with concern about a new viral ‘strain’. I want to explain why such angst is misguided and in the process explain exactly what is a virus strain and a virus isolate.

In science, word usage matters. And sadly, even virologists often do not use their terms properly. I’ve written about it before.

When SARS-CoV-2 is isolated from a COVID-19 patient, that virus is called an isolate. The origin of the term is clear: the virus has been isolated from a patient.

These virus isolates are all the same strain of SARS-CoV-2. They are not different strains, even if they have changes in their genome sequences. A virus strain is an isolate with a different biological property, such as binding to a different receptor, or having a distinctly different stability at higher temperatures, to give just two of many possible examples.

There is only one strain of SARS-CoV-2. The first virus isolate, taken from a Wuhan patient in December 2019, is the same strain as the most recent isolate taken anywhere else in the world in May 2020. So far no one has shown that any of these virus isolates differ in any fundamental property.

I can hear some of you shouting, but isn’t a nucleotide change enough to make a strain? The answer is a resounding NO. Every virus expelled by an infected individual differs from the next by many base changes. It would be foolish and of little utility to call each patient isolate a strain. That term is reserved only for special changes that confer a new property to the virus.

No doubt you have heard reports of different SARS-CoV-2 strains, but I assure you they are likely wrong. Some time ago it was claimed in China that there were ‘L’ and ’S’ strains with distinct pathogenicity in humans. Wrong. You will also hear that there are eight circulating strains of the virus. Wrong. These are all isolates. None have been shown to have a distinct biological property, no matter what the preprints claim.

Most of these claims are in preprints and, if the scientific review process does its job, most of them will simply be reports of new genome sequences with no associated biological changes.

The most recent offender is a preprint claiming that SARS-CoV-2 with an amino acid change in the spike glycoprotein (D614G) increases the transmissibility of the virus. The claim that this amino acid change increases viral transmission is unsubstantiated and likely incorrect. There is no doubt that viruses with the D614G change are emerging in different geographical regions of the world. Until proven otherwise, their emergence is likely due to the founder effect. Let’s say a virus with D614G emerges during replication in a person’s respiratory tract. If viruses with that change infect the next person, and the next, and so on, then the D614G change will predominate. The change is simply a single nucleotide polymorphism of little consequence. It is the noise produced by error-prone RNA synthesis by the virus. Viruses with D614S are simply virus isolates. They are not strains of SARS-CoV-2.

Because of the founder effect, showing that a particular mutation increases viral transmission in humans is very difficult. Many such claims have been made for other viruses in the past, but none have been proven. One that comes to mind is a single amino acid that emerged in the Ebolavirus glycoprotein early in the 2015 West Africa outbreak and was subsequently found in all isolates. No proof emerged that it was not simply a founder effect.

I would also caution that making claims that SARS-CoV-2 is becoming more transmissible ignores the fact that the virus is already exceedingly transmissible among humans. For an amino acid change such as D614S to be positively selected, as opposed to being maintained as a consequence of the founder effect, requires selective pressure. For such an already highly transmissible virus, the nature of such selection pressure is difficult to discern.

Comments on this entry are closed.

  • Pavle 7 May 2020, 5:30 pm

    Great job, as always.
    Question: How likely is that SARS-CoV2 complication (ACE2 => Virus => ACE +++ => Cytokines storm) is only a byproduct of a secondary self-infection that follows the primary infection of the upper respiratory?

  • No Name 7 May 2020, 5:38 pm
  • Daniel 7 May 2020, 6:09 pm

    Thanks for a great post, but I believe the mutation is D614G, not D614S. Typo?

  • Judit 8 May 2020, 4:01 am

    Thank you for your clear, and calm, explanation. I wonder though: do you know what AA changes might make a difference, and how we would know? My knowledge of protein chemistry is pretty limited, I’m curious if someone could speak to that…

  • Rachael Watkins 8 May 2020, 4:19 am

    I’ve read there are emerging isolates with a viral load 270% more …another was a preprint saying there are 19 that can mutate in place. What does that mean for its mutation advancement. Please could you explain. One last thing. Wuhan originally said upon death they would check males to varify if it could be detected in the semn. I’m reading now in America it is indeed there? True or false.
    Thank you so much

  • Dave 8 May 2020, 4:21 am

    @Judit it’s quite difficult to tell, in advance, what amino acid changes might make a significant difference.

    What is certain is that the change would have to be located in one of the regions essential for structure (quite unlikely for a single AA change to affect this in a meaningful way) or function (slightly more likely). What complicates the matter is that structure and function are often closely related.

    I used to work with simulating experiments of protein folding and protein-protein interaction. It’s a highly complex matter and requires significant amounts of experience and computational power. “Results” are often of limited practical use. The way to determine whether a mutation such as D614G increases virulence would usually be retrospective: the change is detected in the wild and studied extensively. If it can be determined that indeed it seems to increase transmissibility THEN scientists would determine what, in detail, about this mutation influenced the outcomes. The results of such a study would still be quite valuable for our understanding of such dynamics.

  • Gigi 8 May 2020, 6:43 am

    It seems to me that the right answer is: we don’t know how many strains of Sars-Cov-2 there are.

    How can one possibly know for sure whether any of the numerous mutations observed has or doesn’t have a phenotypic effect? The changes can be subtle and difficult to observe given the very different conditions in which the virus is spreading. So I guess it is difficult to know either way and it is just as presomptuous to claim that there is only one strain as to claim that there are many.

  • Wayne 8 May 2020, 7:11 am

    Koch’s postulates has never been fulfilled. This whole thing is fake. Ya, I used to call myself a biologist too…

  • Alexander Meitiv 8 May 2020, 8:01 am

    The issue I have with this rant is that author considers “different biological property” a discontinuous, qualitative change. Darwin would disagree. Several nucleotide changes (in some places in the genome) will change binding affinities, perhaps ever so slightly, but with measurable effect. Things like virulence, infectivity, etc. might also change slightly for different isolates. Every infected cell could produce millions of virions and the likelihood that one of them will go on to be more fit and outcompete the infecting isolate is not small. Calling them a new strain might be a stretch, but the biological properties and therefore their epidemiological effects do evolve, usually in the direction of lower virulence and higher infectivity.

  • Ryan Durk 8 May 2020, 8:29 am

    Thanks for writing this. SARS-CoV-2 transmission already seems mostly just limited by person-to-person contact, as opposed to any property of the virus itself. So, it’d be very surprising seeing a limited mutation (D614S) conferring much transmissibility advantage to the virus. I think the most important thing D614S shows is that the virus can mutate at a potential vaccine target and retain high transmissibility.

  • Gail condra 8 May 2020, 8:51 am

    what does judy mikovits say about all this???first documentary on coronovirus by epoch times is quite interesting as well as Ms Mikovits who was treated very badly and railroaded by big pharma and Fauci regarding vaccines… with such biased news media and UN Agenda 21 whom nancy pelosi fully endorced it is very clear the plan is for totalitarian one world govt and agenda21 explains why the democrat party is against borders and wall to keep out drugs and illegal immigrants….with agenda 21 and now a news media that is not reporting or minimally reporting when truth finally reveals a corrupt FBI causing the newly elected president trump to be falsely accused to ruin his presidency and the media is not upset about it and they prefer to parrot the chinese communist party propaganda regarding this virus all your concerns about the virus seem pretty small….there is a bigger enemy but God who created the world and everything in it is greater and by His grace and mercy the evil will be found out…..Remember David and Goliath…..God does not need experts in war or medicine to get anything done. He can take a boy with a stone to bring down the biggest and the badest…..I will put my trust in the Lord who without his mercy none if us could do anything. may the Lord reveal the lies and the false reporting in His perfect time. He will make good come out of everything. Repent and return to the Lord who wants none to parish but all to be saved.

  • PsyMar 8 May 2020, 9:42 am

    Actually, the Lord wants all to be in a parish. The word you wanted is perish.

  • AP 8 May 2020, 10:04 am

    I’ve heard aspirin has antiviral properties, it stops some kind if virus from replication.., will taking an aspirin in the early onset of coronavirus infection make any difference?

  • David 8 May 2020, 10:07 am

    Excellent write-up. Please state your references.

    On the internet, anybody can write anything. Education comes from validating statements with reputable sources.

  • Zach Bjornson 8 May 2020, 10:53 am

    The authors of the D614G paper attempt to rule out a founder effect by looking at genotype frequency over time within geographic areas though:

    > To differentiate between founder effects and a selective advantage driving the increasing frequency of the G clade in the GISAID data, we applied the suite of tools that we had been developing for the SARS-CoV-2 analysis pipeline (Fig. 2, Fig. 3, Fig. S2 and Fig. S3). A clear and consistent pattern was observed in almost every place where adequate sampling was available. In most countries and states where the COVID-19 epidemic was initiated and where sequences were sampled prior to March 1, the D614 form was the dominant local form early in the epidemic (orange in Figs. 2 and 3). Wherever G614 entered a population, a rapid rise in its frequency followed, and in many cases G614 became the dominant local form in a matter of only a few weeks (Fig. 3 and S3).

  • Stacks Eagle 8 May 2020, 11:08 am

    Shoot, you ain’t speaking as to why we have to live under tyranny and stay indoors. People ain’t sick as all that.

  • Joshuah Rainstar` 8 May 2020, 11:18 am

    Thank you for this insightful and concise article. Many are concerned about the potential for an antibody dependent effect to put them at risk of future complications. How would you estimate the timeframe for a new serotype that might plausibly differ enough to exploit ADE to emerge?

  • Nate 8 May 2020, 11:40 am

    Could someone who is a virologist reading this please explain the term ‘variant’ as it applies to viruses? I’ve been using (and have seen a couple of times) the term ‘variant’ instead of ‘isolate’ in this context.

  • Ghosh 8 May 2020, 12:07 pm

    There is convincing proof from the open source data that 614D was outcompeted by 614G isolates, in most geographic region.

    614G certainly has some advantages which needs to be explored. Let’s not downplay this!

  • Greg Weber 8 May 2020, 2:26 pm

    There is a preprint that makes the case that the D614G mutation is biologically significant:

  • Dr. Guy Gordon 8 May 2020, 2:42 pm

    So the problem appears to be the lack of a good itermediary term for populations of the virus that stem from a single founder, but do not rise to the definition of a strain.

    If you have no terms between an Isolate and a Strain, then people will use strain when they mean anything larger than one sample. And eventually the definition of strain will be expanded to include such usage.

    Usage defines the meaning of words. Not dictionaries.

    If you wish the word to retain a strict definition, *just* pointing

  • Dr. James Prego 8 May 2020, 4:37 pm

    I just read the article Zach referenced, and he does have a good point about it.

  • Zephir 8 May 2020, 11:30 pm

    If all viral samples are isolates, how is it possible they form multilevel tree?

  • Narad 9 May 2020, 1:27 am

    “what does judy mikovits say about all this???”

  • N 9 May 2020, 11:30 am

    Any thoughts this lab made hypothesis? I’ve ran a number of the computational analysis and they are reproducible.

  • Special Prospector 9 May 2020, 12:07 pm

    For context, recent review of the complex biology of this/these coronavirus:


  • ThisOldMan 9 May 2020, 8:36 pm

    Difficult to discern? Allow me to help! There is presently a great deal of debate as to whether SARS-Cov-2 can be transmitted only via “respiratory droplets” greater than 20 microns in diameter, which would say the current 6 foot guidelines provide a significant degree of protection, or via “aerosols” (droplets less than 20 microns which quickly dry out). The most highly contagious viruses known, like measles, don’t mind being dried out, and aerosol transmission allows them to infect people 100 feet away. True, coronaviruses have a membrane envelope which tends to make them susceptible to being dried out, but it’s hardly inconceivable that a single mutation could lower this susceptibility significantly. So I frankly think your categorical claims here are overblown. Nevertheless, I wish we knew …

  • Newbie 10 May 2020, 3:02 am

    Virologists need to write a marker paper to define and clear off the difference for the following –

    What is a strain?
    What is a clade?
    What is type?
    What is a subtype?
    What is a haplotype?
    What is a genotype?

    and finally how people should call D614G by the above definitions?

    Please do take help of epidemiologists, computation biologist, data-scientists and molecular biologist to define this nomenclature.

    This will help a Journalist like me!

  • Kottkenessness 11 May 2020, 4:45 am

    Kottke brought me here in the middle of the night with a cocktail. You got me with single nucleotide polymorphisms. People just don’t get it but nobody does yet. Everything is just an opinion. Have fun be safe love the one you’re with or all of them lol.