As predicted, Monday’s post about how the “lab leak” hypothesis for the origin of SARS-CoV-2, the coronavirus responsible for COVID-19, is becoming a conspiracy theory attracted trolls, although it did not go viral. This latter happening is probably fortunate for me. Given the volume of comments that this post got from just 50% or so more traffic than a typical post was more than enough for me to deal with. (Also, because I don’t have advertising, any increase in traffic that boosts my server costs comes straight out of my own pocket.) Be that as it may, given the protean nature of conspiracy theories and their unrelenting nature, it should surprise no one that, on the very same day I published my post on the “lab leak” hypothesis, elsewhere a new wrinkle to the conspiracy theory hit the press in the Wall Street Journal and the Daily Mail. (How appropriate.) The WSJ published an op-ed by Steven Quay and Richard Muller entitled The Science Suggests a Wuhan Lab Leak, while a headline blared from The Daily Mail, all touting how, supposedly, the presence of a “double CGG” sequence (CGGCGG) in the nucleotide sequence of SARS-CoV-2 was supposedly incontrovertible evidence that this coronavirus could not have arisen naturally and had to have been “engineered” in a laboratory (in Wuhan, wink, wink, nudge, nudge):
It wasn’t long before all the other usual suspects started amplifying this story, including stories with headlines like ‘Damning’ science shows COVID-19 likely engineered in lab: experts (from Fox News and the New York Post) and An unlikely genome sequence is evidence that COVID-19 leaked from a lab, two U.S. experts say (the National Post).
You get the idea.
So what is it about this “double CGG” sequence, basically a sequence of nucleotides, specifically CGGCGG, in the genome of SARS-CoV-2 that, if you believe Quay and Muller, is such compelling, slam-dunk evidence that this particular coronavirus could not possibly have arisen naturally through evolution and must therefore be the product of the nefarious Chinese fiddling with coronaviruses in Wuhan? (I am not exaggerating the rhetoric by very much at all here.) And who are Steven Quay and Richard Muller, anyway? Let’s find out!
A not-so-dynamic scientific duo of not virologists
I debated which of the two questions posed above to answer first, because if I point out how utterly unqualified Quay and Muller are to pontificate so confidently about the genomics and molecular biology of coronaviruses, conspiracy theorists will accuse me of ad hominem attacks. However, expertise matters, and neither of them has it in the relevant disciplines. So I consider it entirely appropriate to “consider the source” before I dive into the actual molecular biology, to explain what the CGGCGG sequence is and why what Quay and Muller write about it is nothing more than nonsense based on ignorance of molecular biology, evolution, virology, and genomics.
So who are the members of this not-so-dynamic scientific duo? Let’s start with Richard Muller, who, according to the WSJ byline, is an emeritus professor of physics at the University of California Berkeley and a former senior scientist at the Lawrence Berkeley National Laboratory. Reading that, I couldn’t help but roll my eyes, given my experience with physicists who think they have expertise in the biological sciences, such as cancer and evolution? Does anyone remember Paul Davies and Charles Lineweaver? They’re two astrophysicists who presume to lecture cancer scientists and evolutionary biologists about the “true” origin of cancer and how to treat it. I won’t go into a lot of detail, other than to say that their “discovery” was an idea about carcinogenesis that had been popular among cancer biologists over 100 years ago but found wanting based on accumulating evidence and abandoned. Yet Davies and Lineweaver treated it as some sort of brilliant and radical new insight that those hidebound cancer researchers just couldn’t accept, hence their negative reaction. I wrote about their “atavistic theory” of cancer on two occasions, explaining why it was nonsense not supported by what we know, as did evolutionary biologist P.Z. Myers. He was not impressed, either.
But what about Richard Muller? He has his own Wikipedia entry, which tells me that his areas of expertise are in physics, which is not surprising, including particle physics, mass spectroscopy, and areas of earth science. Interestingly, although apparently he now accepts the science showing that human activity is the main driver of global climate warming, back in the early 2000s he was a “climate skeptic” critical of the science supporting the conclusion of human-caused climate change and accused of “parroting” dubious attacks on the “hockey stick” curve developed by Michael Mann’s group. Later, he became a founder of the Berkeley Earth Surface Temperature Project, which had been intended to test the finding of climate scientists. At the time, Muller’s group was widely praised by climate science deniers, such as Anthony Watts, who said that he was “prepared to accept whatever result they produce, even if it proves my premise wrong.” Spoiler alert: Muller’s results agreed with the conclusions of the dreaded Intergovernmental Panel on Climate Change (IPCC), whose reports sound the alarm about climate change, and he even wrote an op-ed for the New York Times in 2012 portraying himself as a “converted climate skeptic,” even as he couldn’t help but add this qualification:
I still find that much, if not most, of what is attributed to climate change is speculative, exaggerated or just plain wrong. I’ve analyzed some of the most alarmist claims, and my skepticism about them hasn’t changed.
I suppose that it’s not surprising that Muller would become a COVID-19 crank and sign on to the claims about “CGGCGG” indicating a “lab leak” origin for SARS-CoV-2, given his history with climate science. Flirting with science denial is a bad thing, obviously, but somehow managing to come down on the right side of science after doing so is a good thing. However, the fact that Muller was once a “climate skeptic” (a.k.a. denier) suggests that he has a history of being susceptible to certain forms of science denial and conspiracy theory. At the very least, it suggests that he could be a “brave maverick scientist” who relishes being a contrarian. Whatever the case, Muller’s skills and knowledge as a physicist were far more closely aligned to the sort of scientific skillset needed to investigate climate change (and even then they weren’t that well aligned) than they are now to investigate the origin of a virus causing a pandemic. Sadly, his utter lack of expertise in the relevant scientific disciplines didn’t stop him from teaming up with Dr. Quay.
Speaking of Steven Quay, what about that other half of the “CGGCGG” duo? Just Google his name, and you’ll quickly find his website DrQuay.com. A perusal of that website should tell you all you need to know. Dr. Quay is the president and CEO of Atossa Therapeutics, a company that develops and sells contrast agents for MRI and ultrasound and nasal sprays to deliver vitamin B12 and calcitonin. These are all interesting products, but none requires complex knowledge of molecular biology of the sort that leads Dr. Quay to make such definite bold claims about CGGCGG.
Although he doesn’t mention it on his website, Google search results indicate that Dr. Quay is a board-certified pathologist, which might or might not give him some of the skillset needed to analyze a new virus. It would depend upon whether he is primarily a research pathologist and whether his interests involve the appropriate virology and molecular biology needed to analyze the genome of a coronavirus. (Spoiler alert again: He isn’t, and they don’t.) A PubMed search for his publications returns a number of publications, but none that would suggest that Dr. Quay has the necessary expertise to make the claims he makes in his WSJ op-ed.
Then there’s the grift. Yes, Dr. Quay is selling a book about how to survive the COVID-19 pandemic, because of course he is:
It’s advertised this way, with a part about supplements:
So, to boil it down, what we have here is a not-so-dynamic duo of not-virologists, not-molecular biologists, not-geneticists, who think they’ve found evidence of “engineering” in the CGGCGG sequence in SARS-CoV-2 that real virologists, molecular biologists, and geneticists have missed.
They haven’t, but I have to go into the weeds a bit to show why.
What the heck is this CGGCGG sequence, anyway?
Early in the pandemic, after SARS-CoV-2 was isolated and sequenced, scientists noted an interesting and very unusual feature. I went back to this article from May 2020 that identifies what’s so interesting about the CGGCGG sequence, which is in the center of what the article called CCU CGG CGG GCA: The Twelve Letters That Changed the World. This is the furin cleavage site. Does that term sound familiar? It should. I mentioned it in my last post, and it’s a trope that conspiracy theorists have been using for a long time and that was popularized by Nicholas Wade. It’s a site that allows an enzyme in nearly all human cells (furin) to cleave the S1 and S2 segments of the spike protein into their final form. To understand its significance, though, a bit of Biology 101 is required.
DNA and RNA are made up of building block molecules strung together called nucleotides, and the “letters” are abbreviations for the specific nucleotide (cytosine [C], guanine [G], adenine [A] or thymine [T]). RNA is similar to DNA, but differs in three main ways:
- RNA is usually a single-stranded molecule, unlike DNA, which most commonly takes the structure of a double helix. (There are exceptions, such as double-stranded RNA viruses. COVID-19 is a single-stranded RNA virus.)
- The sugar-phosphate “backbone” of DNA contains deoxyribose. RNA contains ribose instead. As a result, RNA is more chemically unstable, which is why messenger RNA half lives are generally short and the reason why the RNA in COVID-19 vaccines had to be chemically modified so as not to be degraded too rapidly.
- The complementary base (the base that pairs to it in a double strand) to adenine in DNA is thymine, whereas in RNA, it is uracil, which is an unmethylated form of thymine.
Proteins are encoded by DNA using the genetic code, in which three “letters” (nucleotides) provide the code for the protein translation machinery of the cell to incorporate one amino acid into a protein:
Thus CGGCGG codes for two arginine residues in a row. I’ll get into the significance of this in a moment. As a preview, I’ll note that the key to these claims is that CGGCGG is so rare in coronaviruses in nature as to be almost certainly due to lab manipulation. It’s not.
Let’s look at what Quay and Muller claim:
A genome is a blueprint for the factory of a cell to make proteins. The language is made up of three-letter “words,” 64 in total, that represent the 20 different amino acids. For example, there are six different words for the amino acid arginine, the one that is often used in supercharging viruses. Every cell has a different preference for which word it likes to use most.
So far, there’s nothing objectionable here. (Actually, I think they mean every organism, not every cell. Codon preference is baked into the genome of an organism.) unfortunately I knew where this not-so-dynamic duo were going:
In the case of the gain-of-function supercharge, other sequences could have been spliced into this same site. Instead of a CGG-CGG (known as “double CGG”) that tells the protein factory to make two arginine amino acids in a row, you’ll obtain equal lethality by splicing any one of 35 of the other two-word combinations for double arginine. If the insertion takes place naturally, say through recombination, then one of those 35 other sequences is far more likely to appear; CGG is rarely used in the class of coronaviruses that can recombine with CoV-2.
In fact, in the entire class of coronaviruses that includes CoV-2, the CGG-CGG combination has never been found naturally. That means the common method of viruses picking up new skills, called recombination, cannot operate here. A virus simply cannot pick up a sequence from another virus if that sequence isn’t present in any other virus.
No, this is not true at all, at least not exactly. First of all, CGGCGG is not all that uncommon. It has been found in other coronaviruses, for example, some isolates of MERS coronavirus. Furin cleavage sites are also found in a number of other coronaviruses, as discussed in this recent review article. Although uncommon, furin cleavage sites are not so uncommon in coronaviruses as to be any sort of strong evidence of laboratory manipulation.
Moreover, there are known mechanisms by which such a sequence could have arisen. This has been discussed extensively on Twitter, including the fact that this nonsense from Quay and Muller is not new and was debunked many months ago:
The CGGCGG sequence could have been caused by copy choice error or frame shift mutation, two very well-established mechanisms taught in basic molecular biology class:
The CGGCGG sequence is also not as uncommon as Quay and Muller claim:
Moreover, as has been pointed out by scientists, it’s not just the furin cleavage site that is unusual about SARS-CoV-2. There are a number of genetic features that, sadly for conspiracy theorists like Quay and Muller, look like nothing more than messy evolution doing its messy thing:
Or, as was described over a year ago (hence the reference to “my esteemed president” Trump):
The chance that someone would make all those changes in a laboratory is “far-fetched but not impossible,” says Nunberg, now director of the Montana Biotechnology Center. “Sadly, while I do not trust the Chinese to be transparent, I trust the conspiracy theorists (including my esteemed president) less. I don’t think anyone knows enough to purposely engineer a new virus that is so successful,” he says. Nunberg points out that viruses are mutating all the time. And a single infected living being can have up to a trillion infectious viral particles from any given virus in its body. “Never underestimate the ability of the virus to adapt,” he says. “It’s the 1,000 monkeys with typewriters who may eventually, at random, type out a Shakespeare play – stuff happens”.
Actually, though, if you look at Quay and Muller’s core argument, it really does come down to an appeal to incredulity. Just because they can’t believe that this CGGCGG could have arisen naturally, they find specious reasons to argue that it didn’t:
Now the damning fact. It was this exact sequence that appears in CoV-2. Proponents of zoonotic origin must explain why the novel coronavirus, when it mutated or recombined, happened to pick its least favorite combination, the double CGG. Why did it replicate the choice the lab’s gain-of-function researchers would have made?
Yes, it could have happened randomly, through mutations. But do you believe that? At the minimum, this fact—that the coronavirus, with all its random possibilities, took the rare and unnatural combination used by human researchers—implies that the leading theory for the origin of the coronavirus must be laboratory escape.
Of course, I just cited examples of how the “novel coronavirus, when it mutated or recombined, happened to pick its least favorite combination, the double CGG. It’s not particularly mysterious, nor is it nearly as remarkable as Quay and Muller claim.
Note the last part though: “Yes, it could have happened randomly, through mutations. But do you believe that?” Quay and Muller admit that this combination could have happened randomly, and so it could. They just don’t believe that it could have, not because the CGGCGG sequence is in and of itself compelling evidence of lab manipulation, but because they want to believe that SARS-CoV-2 was engineered. This leads them to erroneously assert that this “rare and unnatural combination” must mean a laboratory origin. The problem, of course, is that this combination is not unnatural, not necessarily preferred by human researchers, and not as rare as Quay and Muller claim.
Another scientist weighed in on this very issue on Twitter:
Exactly. Random mutations happen all the time. That’s how evolution works.
Meanwhile, our esteemed breast health entrepreneur and our previously climate science-denying physicist (I couldn’t resist borrowing Dr. Maxmen’s phraseology but slightly altering it) continue their appeal to incredulity, this time incredulity that SARS-CoV-2 could have infected humans so efficiently right from the get-go:
Such early optimization is unprecedented, and it suggests a long period of adaptation that predated its public spread. Science knows of only one way that could be achieved: simulated natural evolution, growing the virus on human cells until the optimum is achieved. That is precisely what is done in gain-of-function research. Mice that are genetically modified to have the same coronavirus receptor as humans, called “humanized mice,” are repeatedly exposed to the virus to encourage adaptation.
This is, again, unscientific. I addressed it before when I noted earlier this week that one other claim that keeps popping up is that SARS-CoV-2 appeared “perfectly adapted” to humans as a host. (Wade alludes to this.) This study suggests a pathway by which natural selection in bats could have created a more “generalist” virus, rather than specifically adapted to humans. As I put it last week, SARS-CoV-2 is not “perfectly adapted to humans,” just well adapted enough to infect them efficiently. It did not have to be “optimized” to humans through serial passage in a laboratory.
It also turns out that the furin cleavage site is also not optimal:
Another key feature often cited as evidence of laboratory origin is the furin cleavage site, where the spike protein is cut in half to “activate” viral material for entry into cells. The viruses most closely related to SARS-CoV-2 don’t have this site, but many others do, including other human coronaviruses. The furin site of SARS-CoV-2 has odd features that no human would design. Its sequence is suboptimal, meaning its cleavage by the enzyme furin is relatively inefficient. Any skilled virologist hoping to give a virus new properties this way would insert a furin site known to be more efficient. The SARS-CoV-2 site has more of the hallmarks of sloppy natural evolution than a human hand. Indeed, a timely analysis last year showed convincingly that it is a product of genetic recombination, a natural feature of coronavirus replication and evolution.
No appeal to engineering is necessary.
CGGCGG: The resurrection of an early pandemic gambit
I’ll conclude by noting, as I did last time, that it is certainly possible that the origin of SARS-CoV-2 was a “lab leak” from the Wuhan Institute of Virology. That such an origin is possible, however, does not make it likely or even equally likely as a natural origin. Certainly, from what we know about the virus and its sequence, we can confidently conclude that it is incredibly unlikely that SARS-CoV-2 was “engineered.” It bears the hallmarks, as Angela Rasmussen and Stephen Goldman wrote in the article cited above, of the messiness of evolution, and “rare” combinations can be produced by evolution because evolution provides so much raw material to work on. It’s basically the law of large numbers, incredibly large numbers, that makes even “rare” outcomes not so rare. Scientists with the requisite expertise understand that.
Hacks like Quay and Muller do not, as Joshua Rosenau points out:
This last part is important to note:
Cranks like Dr. Quay:
I saw that 193-page preprint. I will admit that I didn’t read the whole thing. However, what I did read demonstrated to me that Dr. Quay, too, likes to abuse Bayesian analysis, just like the cranks I discussed last week. And, again, just because the most vocal proponents of the lab leak hypothesis are cranks, hacks, and grifters doesn’t mean that it’s impossible that SARS-CoV-2 escaped from a lab. It just means that the arguments in favor of the lab leak hypothesis are currently brain-meltingly stupid, both logically and scientifically.
Quay and Muller don’t help themselves with stuff like this, either:
I do love good sarcasm. I also like spot-on assessments like this;
Again, it is possible that SARS-CoV-2 escaped from a lab, although it is incredibly unlikely that it was engineered and the possibility that it was a natural coronavirus under study at the Wuhan Institute of Virology that escaped is only marginally less plausible and likely. Both of these “flavors” of lab leak hypothesis, though, the escaped engineered flavor and the natural escaped flavor, are far less likely and plausible explanations for COVID-19 than the long-known and long-studied mechanism of zoonotic transfer; i.e., a virus jumping from animal to human. A conspiracy theory can certainly be based on a claim that is possible. Barring compelling evidence making that “possible” explanation the most likely, it still remains a conspiracy theory, and currently the lab leak hypothesis is most definitely a conspiracy theory.