Visual timeline: ‘Proximal Origin’

Published early in the pandemic, an article in Nature Medicine titled “The proximal origin of SARS-CoV-2” — known to many simply as “Proximal Origin” — helped push the lab leak theory of the origins of COVID-19 to the outer fringes by March 2020.

“We do not believe that any type of laboratory-based scenario is plausible,” the analysis stated, which had been read about three million times within weeks of publication. “Our analyses clearly show that SARS-CoV-2 is not a laboratory construct or a purposefully manipulated virus.”

For years, the infectious diseases community had debated whether cutting-edge virology — gain-of-function research that generates more dangerous viruses — could cause a pandemic. Once a pandemic arrived, that debate abruptly ended.  

The virologists had sold the public on an irrefutably natural origin. 

But emails and Slack chat messages acquired through the Freedom of Information Act and subpoenaed by the House Select Subcommittee on the Coronavirus Crisis show that the virologists who wrote the analysis privately expressed concerns that the virus could have emerged from a lab. They also reveal the depth of involvement of some of the most powerful men in science, including former pandemic response leader and National Institute of Allergy and Infectious Diseases Director Anthony Fauci, in shaping the article. Fauci’s institute had funded the Wuhan Institute of Virology through an American research organization called EcoHealth Alliance.

The controversy over “Proximal Origin” represents years of unanswered questions about how a scientific hypothesis that some scientists and intelligence agencies now consider likely was cast as a conspiracy theory — culminating in a committee hearing this summer.  

Four of the five virologists behind “Proximal Origin” — Scripps Institute virologist Kristian Andersen, University of Sydney virologist Eddie Holmes, University of Edinburgh virologist Andrew Rambaut and Tulane School of Medicine virologist Robert Garry — have subsequently coauthored other high impact articles in favor of the theory that the pandemic emerged from the Huanan Wholesale Seafood Market in Wuhan. (A fifth coauthor of “Proximal Origin,” Columbia University virologist Ian Lipkin, has since distanced himself from the group.) In the world’s most prestigious journals and newspapers, these four virologists have cast the evidence for a natural origin as so compelling — so “dispositive” — as to render evidence pointing to the Wuhan Institute of Virology immaterial or the makings of a conspiracy theory.

The virologists’ Slack messages, emails, early drafts, and interview transcripts span 1,096 pages, while emails released through FOIA include hundreds more — yet the controversy over “Proximal Origin” remains unsettled.

Five thousand people have signed a petition accusing the virologists of fraud and calling for the retraction of “Proximal Origin,” including dozens of professionals in science and technology. Yet the virologists and their defenders say their private messages expressing concern about genetic engineering and the Wuhan Institute of Virology simply show a change of their perspective over time — a “textbook example of the scientific method at work.”

“It is easy to sow doubt – to take sentences from here and there in email streams and compare early thinking with later conclusions and presume any change is due to some unspecified pressure rather than a change in the weight or direction of evidence,” the Australian Academy of Science said in an August statement. 

Yet a new timeline by U.S. Right to Know shows that some of the contradictory private and public statements were written on the very same day. The timeline shows that, even as the virologists incorporated core concepts into their paper that were in favor of a natural origin, they simultaneously bemoaned to their coauthors that they were incomplete — even “crap.”

A few examples:

First, the day after the virologists shared an analysis with their funders arguing that engineering SARS-CoV-2 would require “significant amounts of molecular work,” Garry said privately that a graduate student could do it. “Proximal Origin” argues that any genetic engineer would follow computer modeling from the lab of Ralph Baric, a University of North Carolina virologist and frequent collaborator of the Wuhan Institute of Virology, who had mapped out the mutations optimal for binding to human receptors called ACE2. “Proximal Origin” argues that although the virus binds to human receptors optimally, it does not have these predicted mutations, thus is natural. The same day this argument was first incorporated into their analysis, Garry said that lab escape would not be “crackpot…given the [gain-of-function] research we know is happening.” Other Slack messages demonstrate an awareness of other approaches to researching viruses utilized by the Wuhan Institute of Virology and its collaborators.

Second, the virologists wrote in “Proximal Origin” that scientists would have never experimented with a novel virus that had not been published in the scientific literature. Privately the virologists acknowledged that the Wuhan Institute of Virology had “loads [of novel viruses] in their freezers” and that they could conduct engineering experiments on novel viruses “on a whim.” This concept at the core of their paper — that a novel virus would never be subject to gain-of-function experiments — was described by the virologists themselves as partially “foolhardy” and “crap.” They cribbed the idea from one of the world’s most controversial gain-of-function virologists, who was concerned about new biosafety regulations. They described him internally as “unbelievably conflicted.” He was not credited.

Third, the virologists publicly have attributed their rapid shift from favoring a lab origin to favoring a zoonotic origin largely to the news that a coronavirus up to 99 percent similar to SARS-CoV-2 had been discovered in pangolins, scaly anteaters used in traditional Chinese medicine. But the newly released messages show that within minutes of receiving the sequence of this virus, the “Proximal Origin” virologists noticed it was approximately 90 percent similar to SARS-CoV-2, far too dissimilar to serve as a precursor virus. “The more pango sequences I see the less likely I find that they are intermediate – I think they’re just one of many animals with SARS-like CoVs,” Andersen said. Rambaut told the New York Times that a virus identified at the Wuhan Institute of Virology called RaTG13 that is 96% similar to the novel coronavirus was not a close relative. Two days later, Rambaut grumbled to his coauthors that coronaviruses allegedly found in pangolins were not close enough to SARS-CoV-2 to dispel concerns about the Wuhan lab’s cousin virus. “Up to 99% [similar] is no good. There is a 342 [base pair] stretch of RaTG13 that is identical to [SARS-CoV-2]. Sigh,” Rambaut wrote. Two days after the pangolin coronavirus sequence was made available, Andersen and Rambaut said privately that they remained undecided between a natural or lab origin.

Fourth, the virologists made jokes about the possibility that the Wuhan Institute of Virology and its American partner EcoHealth Alliance ignited the pandemic, even as they drafted an analysis that helped shield these institutions from greater scrutiny. An early draft of “Proximal Origin” dismissed as “conspiracy theories” concerns about an American collaboration with the Wuhan Institute of Virology to conduct gain-of-function research on coronaviruses. The same day this idea was incorporated into the draft, Rambaut joked that perhaps EcoHealth Alliance “had planned a press conference predicting which virus would cause the next pandemic but then it escaped from the lab early.” The same day a preprint of “Proximal Origin” published largely dismissing any lab scenario, Andersen shared privately that the suspicions of Sen. Tom Cotton, R-Ark., about the Wuhan Institute of Virology were on target. “Tom Cotton is trending with COVID-19 on the Twitters at the moment. I gotta say – the guy isn’t totally wrong,” he said.

In addition, the newly revealed messages raise questions about the integrity not only of “Proximal Origin” but other high impact papers the virologists would go on to coauthor in favor of the zoonotic hypothesis. 

The virologists would privately justify asserting a natural origin with limited genomic evidence because of the epidemiological evidence that the virus had emerged from live animal stalls at a wet market. But the new private messages show that they were not confident in that evidence either, expressing skepticism about the quality of data out of Wuhan, data they would go on to employ to assert a wet market origin with near certainty in other prestigious outlets like Science and The Atlantic. 

The ‘bizarre backstory’ of ‘Proximal Origin’

The genesis of “Proximal Origin” is by now well known: The virologists expressed alarm about a section of the SARS-CoV-2 genome that was “inconsistent with expectations from evolutionary theory.”  After a teleconference on February 1, 2020, with some of the world’s most influential funders of scientific research — Wellcome Trust Director Jeremy Farrar in the U.K., and National Institutes of Health Director Francis Collins and Fauci in the U.S. — they began accumulating evidence against that theory that the virus could have originated in a lab. 

The purpose of the call was described by one of the virologists as “challenging a certain theory and if we could, drop it.“

The virologists’ messages show that their article was written under the direction of Farrar, who served as unofficial liaison between the virologists and Fauci and Collins. 

The incentive to produce something these “higher ups” would like appears to have been  immense. Altogether Collins, Fauci and Farrar oversaw about $72 billion in scientific funding that year.

Despite the difficulties posed by Beijing’s censorship and the novel coronavirus’ unexplained properties, the virologists wrote a paper within the span of a month that could serve as a “conclusive statement which will then be the go to scientific statement to refer to,” according to Farrar.

Farrar, who had acquired a burner phone for the purpose of discussing the possibility the virus had originated from a lab, also simultaneously shepherded drafts to the World Health Organization, emails show. Today Farrar serves as chief scientist at the WHO.

The “Proximal Origin” analysis was also shaped by controversial gain-of-function virologists concerned about restrictions on future research, and watered down in response to editors at the prestigious scientific journal Nature Medicine who sought a conclusive statement against any lab origin.

Soon after the call Farrar, Fauci, and Collins, the virologists began aligning their analysis to the conclusion predetermined by many within the scientific community.

“I personally think we should get away from all the strange coincidence stuff. I agree it smells really fishy but without a smoking gun it will not do us any good. The truth is never going to come out (if escape is the truth),” Rambaut said the day after the teleconference. “Given the shit show that would happen if anyone serious accused the Chinese of even accidental release, my feeling is we should say that given there is no evidence of a specifically engineered virus, we cannot possibly distinguish between natural evolution and escape so we are content with ascribing it to natural processes.”

“I totally agree that that’s a very reasonable conclusion. Although I hate when politics is injected into science – but it’s impossible not to, especially given the circumstances,” Andersen replied.

Three days after the teleconference, Andersen advised the National Academies of Sciences, Engineering and Medicine that it should come out more forcefully against the “crackpot” and “fringe” theory that SARS-CoV-2 had been engineered as a bioweapon or for basic research, and that the virus is “consistent with natural evolution,” precisely reversing what he said privately a couple of days earlier. (“We all find the genome inconsistent with expectations from evolutionary theory. … I’m pretty sure by now they think I’m a complete crackpot. … It’s [lab emergence] not some fringe theory.”) 

The scientists behind “Proximal Origin” expressed a sense of collaboration with these powerful scientific funders to dampen the global debate.

“Pressure from on high … all came together very quickly in the end. Jeremy Farrar and Francis Collins are very happy. Works for me,” University of Sydney virologist Eddie Holmes said.

The virologists searched for natural explanations for the furin cleavage site — a feature on the spike protein that supercharged the novel coronavirus into the most widespread pandemic pathogen in a century. They grasped for a natural explanation for the immediate transmissibility of SARS-CoV-2 as compared to other coronaviruses like SARS or MERS.

But it wasn’t easy. 

Their private messages describe pangolin coronavirus data as “sketchy” and too dissimilar to SARS-CoV-2 to be meaningful, acknowledged that a natural origin would require a “convoluted” series of events, noted the “weird” suppression of key data by Beijing authorities, and grappled with an “iffy” and “majorly confusing” viral family tree. 

After weeks of digging for evidence, the virologists concluded that the unusual features of the novel coronavirus could occur in nature.  A draft of their paper stated that a lab origin was “not necessary,” but the published version went further, asserting that a lab origin was not “plausible.” 

Many messages suggest the purpose of the article was to steer the media, the public, and policymakers away from speculation about lab-related scenarios.

While the group expressed some concerns about rumors on the fringes — about deliberately released bioweapons and a withdrawn paper about SARS-CoV-2 containing bits of HIV — the recently revealed messages also show the virologists helped to quash credible concerns emanating from the scientific community and U.S. government. 

Marc Lipsitch, a Harvard University epidemiologist, had circulated concerning comments raised about the lab in Wuhan by Richard Ebright, a molecular biologist at Rutgers University — both longtime critics of the riskiest gain-of-function research. The “Proximal Origin” virologists were fielding questions from reporters at Science and the New York Times.  

The influence of ‘Proximal Origin’

The article was endorsed by Fauci in front of reporters and cameras at the White House, cited by the Defense Intelligence Agency, and briefed to analysts at the State Department. According to recent reporting, Fauci cited the “Andersen paper” to persuade the Central Intelligence Agency to alter its assessment of the origin of the pandemic from lab to nature.

Since its publication, the virologists behind “Proximal Origin” have received awards, appeared in newspapers around the world, and have continued to highlight their authorship of one of the most widely cited and read scientific papers in history. “Proximal Origin” was published in the spring of 2020 and by that summer, Fauci’s NIAID had awarded $82 million to a network of scientists that included Andersen, Garry and EcoHealth Alliance. Yet despite all of their professional success, the virologists have yet to quiet all of the questions about their integrity. 

Pressed by members of Congress at a hearing this summer, Andersen, flanked by a team of attorneys, backed off prior claims that anyone who disagreed with “Proximal Origin” was a conspiracy theorist.

But in the months since, Andersen has deflected criticism about “Proximal Origin” by returning to a familiar refrain.

“Accusing domain experts of fraud is a commonly used tactic by conspiracy theorists,” he said in a tweet.

A faulty premise

On January 21, 2020, Chinese scientists highlighted a feature of SARS-CoV-2 called a furin cleavage site situated in an ideal place in the spike protein of the novel coronavirus — a feature never seen before in a coronavirus of its kind. 

Furin cleavage sites had been shown to make flu viruses more dangerous, and would have been easy to add “without a trace.”

The “Proximal Origin” virologists predicted it would endow the new virus with significantly stronger transmissibility than SARS. The discovery put them on edge. 

Days later, the Wuhan Institute of Virology released a preprint announcing the lab had identified a close cousin to the novel coronavirus in its freezers — a 96 percent match. 

These two revelations were largely responsible for igniting the chain of events that led to “Proximal Origin.” 

On a February 1 conference call, Andersen and Holmes presented their concerns about the furin cleavage site and the Wuhan Institute of Virology. Present on the call were Farrar, Fauci, Collins, all of the “Proximal Origin” authors except for Lipkin, and three virologists looped in by Farrar — Erasmus Medical Centre virologists Ron Fouchier and Marion Koopmans, and Charité virologist Christian Drosten. 

Fouchier and Drosten — who have both conducted gain-of-function experiments and were considered “conflicted” — pushed back forcefully on the hypothesis that SARS-CoV-2 was engineered.

They argued that labs would never pick a completely novel coronavirus as a “backbone” for gain-of-function experiments. 

Their colleagues delivered their rebuke in the presence of the virologists’ funders, Farrar, Fauci and Collins. The presence of Fauci and Collins may have been particularly sensitive given their history of support for the riskiest gain-of-function research. Fauci’s institute at the NIH had funded coronavirus discovery and gain-of-function experiments at the Wuhan Institute of Virology, records would later show, something he had been made aware of a few days earlier.

But notes written by Andersen on the same day as the call show that Fouchier’s and Drosten’s opinions didn’t dampen the authors’ concerns about engineering. 

“The only thing that remains perplexing … is the fact that it has a furin site with O-linked glycans in the spike protein between S1 and S2,” the notes read.

Soon after the call, four of the five authors of “Proximal Origin” — everyone but Lipkin — began exchanging notes on Slack. Andersen named the chat “Project Wuhan Engineering.”

“Nice channel title,” Rambaut said.

“Super secret too,” Andersen replied. “I think the main thing still in my mind is that the lab escape version of this is so friggin’ likely to have happened because they were already doing this type of work and the molecular data is fully consistent with that scenario.”

In the more private forum, Andersen commented on the pushback from Fouchier and Drosten. A cursory review of work in Wuhan showed the lab’s interest in discovering novel bat coronaviruses that had not yet emerged in human populations and testing which genetic mutations could transform them into dangerous viruses. 

“Both Ron and Christian are much too conflicted to think about this issue straight – to them, the hypothesis of accidental lab escape is so unlikely and not something they want to consider. The main issue is that accidental escape is in fact highly likely – it’s not some fringe theory,” Andersen wrote the next day, on February 2. “I absolutely agree that we can’t prove one way or the other, but we never will be able to.”

“That doesn’t mean that by default the data is currently much more suggestive of a natural origin,” he continued. “It’s not – the furin cleavage site is very hard to explain.”

“My fear is that the likes of [Drosten] and [Fouchier] put the question that’s being asked here into the same category [as conspiracy theories] … I’m pretty sure by now they think I’m a complete crackpot,” Andersen said.

Andersen said that engineering would be “impossible to prove” except for “the most simple scenario.” Yet this “simple scenario” would be described as emblematic of nearly all plausible virology work involving engineering in “Proximal Origin.”

“He is looking for very specific evidence proving this is unnatural (which is understandable), but except for the most simple scenario where somebody plugged a gene into a preexisting backbone, [engineering] would simply be impossible to prove,” Andersen said. “The presence of furin a priori moves me slightly more towards accidental release, but it’s well above my paygrade to call the shots on a final conclusion.”

Garry also appeared to challenge the notion that a lab would never work with a novel virus in an email to his coauthors on February 2. Biological engineers would in fact be likely to study a bat virus that had not yet spilled over into humans, he said.

“If you were doing gain of function research you would NOT use…SARS or MERS…These viruses are already human pathogens,” he said. “What you would do is choose a bat virus that had not yet emerged [in humans].” 

Garry continues by describing “simple” hypothetical experiments involving engineering and serial passage of coronaviruses that could have resulted in SARS-CoV-2. This sort of experimentation would eventually be dismissed as implausibly complicated in the published version of “Proximal Origin.” 

“I aligned [the novel coronavirus] with the 96% bat CoV sequenced at the [Wuhan Institute of Virology]. Except for the RBD the S proteins are essentially identical at the amino acid level – well all but the perfect insertion of 12 nucleotides that adds the furin site. S2 is over its whole length essentially identical. Do the alignment of the spikes at the amino acid level – its stunning. Of course, in the lab it would be easy to generate the perfect 12 base insert that you wanted,” he said.

“A likely explanation could be something as simple as passage SARS-live CoVs in tissue culture on human cell lines (under BSL-2) for an extended period of time, accidentally creating a virus that would be primed for rapid transmission between humans via gain of furin site (from tissue culture) and adaption to human ACE2 receptor via repeated passage,” Garry said. “I really can’t think of a plausible natural scenario where you get from the bat virus or one very similar to it to nCoV where you insert exactly 4 amino acids, 12 nucleotides, that all have to be added at the exact same time to gain this function – that and you don’t change any other amino acid in S2? I just can’t figure out how this gets accomplished in nature. In this scenario RaTG13 the WIV virus was generated by a perfect deletion of 12 nucleotides while essentially not changing any other S2 amino acid. Even more implausible IMO.”

“Maybe then pass [the novel bat coronavirus] in human cells for a while to lock in the [receptor binding domain], then you reclone and put in the mutations you are interested in — one of the first a polybasic cleavage site,” he continued, using another term for the furin cleavage site. 

“The furin site would be the first thing to add for sure,” Andersen said. 

Fouchier and Drosten also underestimated the lab’s capabilities, Andersen said.

“For the SARS [gain-of-function] studies they created a reverse genetics system for their bat virus on a whim,” Andersen said. “So Ron and Christian’s argument (which I found to be the strongest) about that not being feasible is not true – they were already creating those.”

The virologists further questioned Fouchier’s impartiality.

“Ron … is not going to want it to be a GOF [gain-of-function] escape,” Rambaut said.

Fouchier’s name in the scientific community has become synonymous with controversial research on highly pathogenic avian flu — which is estimated to have a 50 percent fatality rate — generating a new airborne virus. The New York Times described it as the “doomsday virus.” Fouchier’s experiment spurred a debate in the scientific community about whether scientific experiments with the potential to cause pandemics should be permitted.

Despite their private skepticism and concerns about Fouchier’s conflicts of interest, his ideas were immediately incorporated into the earliest drafts of “Proximal Origin.”

“The virus backbone (beyond spike) is not an indicator of a human source of 2019-nCoV emergence,” Fouchier wrote in a Feb. 2 email to the group. “The virus itself has not been described or characterized previously and no reverse genetics system has been described for this virus.”

It became the paper’s keystone argument against the hypothesis that SARS-CoV-2 was engineered. 

“If genetic manipulation had been performed, one of the several reverse-genetic systems available for betacoronaviruses would probably have been used,” the article echoed. “However, the genetic data irrefutably show that SARS-CoV-2 is not derived from any previously used virus backbone.”

Just three days after voicing his concerns, Andersen adopted Fouchier’s argument in a conversation with a group of scientists tapped by the National Academies of Sciences, Engineering, and Medicine to examine COVID-19’s origins for the White House Office of Science and Technology Policy — a group that Andersen said he believed was organized by Fauci. The Office of the Director of National Intelligence and the Federal Bureau of Investigation were also involved.

“I do wonder if we need to be more firm on the question of engineering. The main crackpot theories going around at the moment relate to this virus being somehow engineered with intent and that is demonstrably not the case,” Andersen said.

“Engineering can mean many things and could be done for either basic research or nefarious reasons, but the data conclusively show that neither was done (in the nefarious scenario somebody would have used a SARS/MERS backbone and optimal ACE2 binding as previously described, and for the basic research scenario would have used one of the many already available reverse genetic systems),” he said. “If one of the main purposes of this document is to counter those fringe theories, I think it’s very important that we do so strongly and in plain language (‘consistent with’ [natural evolution] is a favorite of mine when talking to scientists, but not when talking to the public – especially conspiracy theorists).”

By this date, the draft dismissed the virus “being engineered or otherwise created with intent” as “largely incompatible with the data.”

The day after his email to the National Academies stating the virus is “consistent with natural evolution,” and two days after drafts of “Proximal Origin” began dismissing the idea of a virus “created with intent,” Andersen told his coauthors he leaned toward the virus having been developed in cell culture in the lab. 

“Definite lean for me too,” Garry replied.

Serial passage 

After discounting engineering, Andersen and Garry said he leaned toward a lab origin through a technique known as serial passage. This technique can generate more dangerous viruses in the lab through serial infections in cells or lab animals.  Andersen and Garry appeared to think it could be the most likely explanation for the unusual features of SARS-CoV-2.

On February 8, an early draft of “Proximal Origin” circulated to the scientific funders — Farrar, Fauci, Collins — and the gain-of-function virologists — Fouchier, Koopmans and Drosten — firmly ruled out the possibility that SARS-CoV-2 had been engineered, but still considered a lab origin via serial passage a viable scenario. The group also considered spillover through an intermediate host and undetected circulation in humans.

The changes did not go far enough for Fouchier. In response to the new draft, he once again pushed back on any lab origin by arguing no close relative to SARS-CoV-2 had been described in published scientific literature.

“I am not in favor of publishing as is. I fail to see how the last of the three discussed scenarios (passaging) does not fall under the category of ‘laboratory manipulation.’ There is no evidence that might hint to this scenario and hence it should be put aside just like the engineering option,” he said. “As far as I am aware, no laboratory has worked on passaging the pangolin-origin virus, the bat-CoV RaTG13, or another closely related virus or had access to it prior to the outbreak. That nCoV-2019 could originate from a SARS-like virus in Chinese labs can also be excluded.”

The authors did not buy that argument either.

“Crap comments … basically just saying it can’t be true,” Holmes said.

“Not only has this been done, it’s specifically being done in Wuhan. In BSL 2,” Andersen wrote. “That in itself means that we can’t just dismiss a lab theory off hand by saying ‘not possible.’ That would be very foolhardy.”

“The furin link keeps bugging me too,” Andersen continued. “I can’t find any good references on it in the published literature for [coronaviruses].”

“Conflating the absence of evidence with actual evidence against,” Rambaut said.

Fouchier also said that he opposed a draft of the article that considered the lab leak theory because it could lead to “harm to science”through renewed debates about biosafety.

“This manuscript would be much stronger if it focused on the likelihood of the first two scenarios as compared to intentional or accidental release,” Fouchier said. “That would also limit the chance of new biosafety discussions that would unnecessarily obstruct future attempts of virus culturing for research and diagnostic purposes for any (emerging/zoonotic) virus.”

Koopmans, the director of Fouchier’s department at Erasmus Medical Center, opposed calling attention to the furin cleavage site at all.

“By zooming in on a specific finding that is NOT in the public domain as far as I know, I think this will generate its own conspiracy theories,” Koopmans said.

“Call me conspiratorial (OK that horse left the barn), but I think there may be some hallway talk going on at Erasmus,” Garry said.

“Unbelievable how conflicted Ron is,” Andersen grumbled on Slack. “Much too conflicted to think about this issue straight,” he said in a separate message.

The “Proximal Origin” virologists did not immediately discount serial passage in response to this feedback, but did seek other evidence they could use to rule it out. They expressed optimism that a coronavirus found in pangolins would provide the missing pieces.

Andersen reassured Fouchier and Koopmans that he did not support publishing the “open-ended” version of the article they had written without additional evidence for a natural origin. He shared their concern that it could “backfire.”

“As to publishing this document in a journal, I am currently not in favor of doing so,” he said. “I believe that publishing something that is open-ended could backfire at this stage. I think it’s important that we try to gather additional evidence – including waiting on the pangolin virus sequences and further scrutinizing the furin cleavage site and O-linked glycans – before publishing. That way we can (hopefully) come out with some strong conclusive statements that are based on the best data we have access to. I don’t think we are there yet.”

“Let’s hope more [pangolin] sequences come out … Would be critical evidence against cell culture hypothesis (which I’m still leaning towards),” Andersen said on Slack a few days later.

The virologists speculated that prematurely dismissing serial passage in the lab could invite backlash.

“I think we need a stronger argument than an assertion that no lab has done those experiments,” Rambaut said. “The pangolin virus that was announced in the press conference might solve this issue if it has the furin cleavage site insertion which would be all but conclusive for the natural scenario.”

Rambaut expressed hope that the pangolin coronaviruses would provide the proof they needed.

However, when the sequences of the pangolin coronaviruses were published, the virologists’ hopes were dashed — the pangolin coronaviruses lacked a furin cleavage site. 

The “Proximal Origin” virologists reached back to Fouchier’s argument – the one they had dismissed as “crap.” 

“A hypothetical generation of SARS-CoV-2 by cell culture or animal passage would have required prior isolation of a progenitor virus with very high genetic similarity, which has not been described,” the article states. “We do not believe that selection during passage, or any other type of laboratory-based scenario, is plausible.”

No smoking gun

An overarching premise of “Proximal Origin” is that the genome of SARS-CoV-2 alone can disprove the theory the virus is man-made, but many of the virologists involved in “Proximal Origin” made private statements acknowledging lab viruses can be indistinguishable from natural ones.

Garry noted the furin cleavage site appeared inserted “like a scalpel” and that it would be “relatively easy” to insert. 

“Even though there’s some strange research going on here, there’s no smoking gun. Not quite sure what such a gun would look like though,” Andersen said on February 2. “I’m not quite sure what a virus from culture vs. an intermediate host would look like – I think they’d probably be indistinguishable.” 

“You can synthesize bits of the genes de novo with perfect precision and then add them in without a trace,” Garry wrote in a February 6, 2020, message. 

Even after “Proximal Origin” was published, Andersen expressed doubts to his coauthors that they could truly rule out engineering since a lab technique called Gibson assembly does not leave signs of genomic manipulation . 

“No obvious signs of engineering anywhere, but that furin site could still have been inserted via Gibson Assembly … and clearly creating the reverse genetic system isn’t hard,” he said.

But he found little support among his coauthors, whose position against the lab leak theory had solidified.

The ‘half furin cleavage site’

The virologists hoped for more evidence of a natural origin so that they could produce the “conclusive” statement sought by Farrar, Fauci and Collins, Fouchier, Koopmans and the journal Nature.

On February 24, Holmes relayed some promising news. His lab had discovered RmYN02, a bat coronavirus in Yunnan Province that was 93 percent similar to SARS-CoV-2. 

Holmes shared unpublished data from his lab showing that RmYN02 had an apparent insertion in the same place in the genome where the SARS-CoV-2 furin cleavage site was located. 

The furin cleavage site in SARS-CoV-2 is made up of twelve nucleotides that encode four amino acids, PRRA. The bat coronavirus has two amino acids in common with the SARS-CoV-2 furin site – PAA.

The new data suggested that it was theoretically possible for a furin site to develop naturally. The virologists called it a “half furin cleavage site.”

They hypothesized that recombination — the exchange of genetic material between at least two separate viral genomes — could explain the furin cleavage site in SARS-CoV-2.

“I’m now very strongly in favour of a natural origin,” Holmes said. “The component bits of the virus are more or less there in a tiny sample of wildlife.”

“I was always in favour of the pre-adapted jump from animals hypothesis but now it is plausible that that was directly from bats,” Rambaut said.

Both Garry and Andersen remained more cautious.

“No polybasic site, HOWEVER, this provides a mechanism,” Andersen wrote.

“I don’t see how it gets us any closer to discriminating between any of the models. There still needed to be recombination and evolution in either an animal, animals, humans or all of the above. It does not rule out or in lab passage,” Garry said, adding that it is “no smoking gun.”

“I think this lends pretty strong support for an animal origin of the ‘confusing’ features of the virus, so I think it’s important to include. None of this disproves accidental lab infection,” said Andersen. “However, it shows that all the steps can occur in nature.”

“Makes it much more likely the full furin site could have been acquired very early in humans or potentially in an intermediate host,” Andersen said.

“It was necessary to examine the lab hypothesis, but we did and it’s not necessary to invoke lab escape and the events leading to nCov-19 all could have and in all likelihood did occur in nature,” said Garry.

But after examining the nucleotides that made up the amino acids, enthusiasm faded.

“You need 2 transitions and three deletions (or insertions) to go between [the furin cleavage sites],” Rambaut said. “l am not convinced these are related inserts. … I still think that all it tells you is there are some bat viruses with an insertion at this site.”

“Yes, but I think that is an enormous given that 99% of the lab escape idea from genomics was the [furin] cleavage site insertion and we’ve not seen this in any other bat virus… I also think it may be a different insertion, but it means these insertions are happening in nature,” Holmes replied.

“I don’t think this data necessarily argues against accidental release, it shows insertions at this site can happen in nature, making the need to reach for a non-natural explanation much diminished,” Andersen said.

“Proximal Origin” would go further than Garry and Andersen’s statements that a lab origin was “not necessary,” asserting a lab origin was not “plausible.”

While some virologists agree a recombination event involving a virus like RmYN02 could explain the furin cleavage site in SARS-CoV-2, other scientists have raised doubts. 

When first asked about his 180 degree turn on the possibility of a lab origin, Andersen said in an email to the New York Times that “features in SARS-CoV-2 that initially suggested possible engineering were identified in related coronaviruses.” 

The virologists’ private messages show that with regard to the furin cleavage site, that’s only “half” true.

‘The paradox’: Holes in the natural origin story

Even after the virologists set aside the mysterious furin cleavage site, holes in the natural origin story remained.

The virus emerged with an infectiousness not seen in any coronavirus in natural history. The receptor binding domain — the segment on the spike protein that binds to host cells — appeared to be highly fit.

“Let’s not forget that what we’ve observed is completely unprecedented as far as I know. Never before has a zoonotic virus jumped into humans and spread through the population like wildfire with this kind of speed,” Andersen said on February 10. “This in itself would require further inquiry.”

The virologists flagged the immediate facility with which the virus infected human cells as among the unusual features of the virus that were “already noticed or easy to discover.” 

Researchers would by February 18 estimate that SARS-CoV-2 bound to human cell receptors called ACE2 between 10 and 20 times better than SARS.

The virologists at once acknowledged that the novel virus appeared “optimized” to bind to ACE2, yet argued that “importantly, the interaction is not predicted to be optimal” by previous computer modeling.

The virologists argued that any genetic engineer would use a model of optimal receptor binding discovered in 2008 by a frequent collaborator with the Wuhan Institute of Virology — University of North Carolina coronavirus engineering expert Ralph Baric. 

They described this observation as “strong evidence” the virus is not lab-made. 

By February 3, the earliest drafts of “Proximal Origin” included the argument that a scientist interested in working with coronaviruses would exclusively rely on such modeling for his experiments.

“While these analyses suggest that 2019-nCoV may be capable of binding the human ACE2 receptor with high affinity, importantly, the interaction is not predicted to be optimal,” a February 3 draft reads. “Additionally, several of the key residues in the RBD of 2019-nCoV are different from those previously described to be optimal for human ACE2 receptor binding. This latter point is strong evidence against 2019-nCoV being specifically engineered as, presumably, in such a scenario the most optimal residues would have been introduced, which is not what we observe.”

The same day, Garry sent a Slack that appears to contradict the idea that a genetic engineer would only follow the path determined by this computer modeling.

“It’s not crackpot to suggest [lab escape] could have happened given the GoF research we know is happening,” Garry said on February 3.

The argument in “Proximal Origin” also clashes with statements the virologists made suggesting that the literature published by the Wuhan Institute of Virology and the Baric lab appeared to be almost a “how-to manual” for constructing SARS-CoV-2 in the lab.

The argument in “Proximal Origin” only considers one narrowly defined lab-based scenario. Yet the virologists privately discussed other approaches that could have led to the generation of a virus like SARS-CoV-2 in a lab — through reverse genetics, or a combination of engineering and serial passage —  approaches that the authors privately acknowledged the Wuhan lab to have the capacity and expertise to undertake.

Other messages make clear that even after incorporating this argument into their analysis, the virologists remained puzzled by how the virus had emerged preadapted for human spread.

Privately Andersen on February 11 described the “optimal human ACE2 [receptor binding domain]” as “not so consistent” with a natural spillover of the virus from an animal to a human. 

The spike protein’s evolution as measured by a ratio called dN/dS also posed a problem. Andersen found early SARS-CoV-2 spike sequences had a low dN/dS, showing it wasn’t changing much. This calculation suggested the spike had emerged highly evolved to infect human cells. This contrasted with SARS, whose dN/dS was initially high, and then decreased as it circulated and adapted to humans.

“It is indeed striking that this virus is so closely related to SARS yet is behaving so differently. It seems to have been preadapted for human spread since the get go. It’s the epidemiology I find most worrying,” Holmes said in a message on February 10. 

“This whole thing is doing my brain in. I literally swivel day by day thinking it is a lab escape or natural,” Rambaut said on February 20, several days after supposed new evidence led the group to do a 180-degree turn on their initial concerns about a possible lab origin. 

“My brain is a badly calibrated [Markov chain Monte Carlo],” Andersen agreed.

The evidence that SARS-CoV-2 emerged instantaneously pandemic-ready also posed a “paradox.” 

In drafts of the paper dated February 14 and February 15, the virologists note “a paradox,” but note they are “not sure whether to include” it. 

“We still have the paradox,” Rambaut said on February 24 as they prepared their manuscript for publication. “If the virus is human adapted, it should have started circulating as soon as it arose. But we don’t see any genetic variants that are likely older than Autumn 2019.”

“Yes – paradox still in full force,” Garry replied.

This inconvenient complexity was ultimately omitted from the analysis.

While a “half furin cleavage site” had been identified, the bat coronavirus with the “half furin cleavage site” had a very different receptor binding domain than the virus’ closest relative, RaTG13, as well as the pangolin coronaviruses. How these viruses or related viruses could have recombined into SARS-CoV-2 remained perplexing. 

“Some convoluted shit going on here,” Rambaut said.

On February 26, as the manuscript was finalized, the virologists decided to avoid “proposing anything too specific.”

“I think we should stick to the original plan for this article as much as possible and not try to be too detailed about what we think happened (e.g. which bits in which hosts),” Holmes said. “I say this because I am certain that the picture is going to change rapidly as new data come out and I am loathe to make any strong conclusions when the sample is so small.  … I just don’t think we need to propose anything too specific.”

Even in skirting these complex questions, the virologists were confronted with a lab scenario that was impossible to rule out.

Genomic analysis couldn’t rule out the scenario of a researcher becoming accidentally infected by a natural virus  — such as the dozens of SARS-like coronaviruses the Wuhan Institute of Virology had collected over the years.

“At this stage we unfortunately just can’t rule out a potential accidental infection from the lab,” Andersen said.

“No, we can’t and should not because that would have precipitated the cries COVER-UP. No doubt Tulane would have been implicated,” Garry replied.

But ultimately the virologists decided they “shouldn’t even mention it.”  

“The only thing that is left in the ‘conspiracy’ side of things is that a researcher became infected through handling, sampling or culturing bat viruses (i.e. the exact one that became nCoV),” Rambaut said on February 25. “But we don’t (and cannot) address the actual nature of the zoonotic event from an evolutionary/genomic event so we shouldn’t even mention it.”

The ‘spurious’ connection to the wet market

The virologists acknowledged amongst themselves that the data pointing to a wet market may be shaky or the output of a censorious regime.

These statements contrast starkly with the public stances they would become famous for in scientific papers and the press in the following years. The coauthors of the “Proximal Origin” analysis would go on to lead the world’s most headline-grabbing scientific analyses in favor of the wet market theory of the pandemic’s origins.

The virologists speculated whether the wet market outbreak was a “red herring” due to “detection bias,” which would artificially make it seem like Covid came from the market simply because authorities mainly looked for cases connected to the market at the beginning of the pandemic instead of looking at other areas in the city.

“Someone should tell Nature that the fish market probably did not start the outbreak,” Garry said, as the virologists weighed the possibility of undetected circulation in humans prior to that outbreak.

On January 24, 2020, a paper in The Lancet found 13 of the 41 first official cases had no connection to the market. The “Proximal Origin” authors said in their earliest notes that the wet market connection may be “spurious” and that there was “some doubt on that already.”

While furin cleavage sites had been studied in highly pathogenic avian flu, it generally required very crowded and poorly ventilated conditions to arise. The wildlife sold at the Huanan Wholesale Seafood Market were not packed densely enough to facilitate the evolution of a furin cleavage site.

“The hypothetical [progenitor] virus with the optimal furin like site also had to evolve a near perfect RDB that was as [Andersen] put it ‘lock and loaded’ to bind to human ACE2,” Garry said on February 3. “The major hangup I have is the polybasic cleavage site. Clearly it can arise in highly pathogenic avian flu, but it’s not really a ‘natural’ process.”

“No way the selection could occur in the market. Too low a density of mammals: really just small groups of 3-4 in cases,” Holmes said.

“That is what I thought as well, which begs the question where would you get intense enough transmission (like the poultry farms for [H5N1]) to generate and pass on the furin site insertion?” Garry replied.

“That’s the million dollar question,” Rambaut said.

The virologists expressed concern about the clampdown on genomic data and other epidemiological evidence from the wet market. 

Yet the same virologists would set aside these concerns in subsequent years to assert the certainty of a wet market origin.

In a 2022 paper in Science whose coauthors included Holmes, Andersen, Garry and Rambaut, the virologists concluded that the official case numbers pinpointed the wet market as the pandemic’s epicenter and rejected concerns that biased data collection by the China Centers for Disease Control skewed this picture.

Like “Proximal Origin,” the Science paper gave the impression that its conclusions were airtight and that the origins debate had concluded. Like “Proximal Origin,” it received global media coverage.

But privately, Holmes said that official case numbers from China may not be accurate.

On March 29, 2020, about two weeks after “Proximal Origin” was published, Holmes said that according to his sources in the country, COVID-19 cases in China outstripped officially reported cases several times over, and that some hospitals in Wuhan had been discouraged from reporting cases.

“There is so much repression and deceit it is ridiculous,” Holmes said. “The true number of cases is probably a log more than reporting (I was consistently hearing 5% prevalence in Wuhan). I’ve also heard some of the hospitals in Wuhan are declining to test because they want to report low/no numbers.”

The 2022 Science paper further reported that positive samples swabbed around the wet market clustered where live animals were sold, showing that the pandemic emerged from a zoonotic jump. 

But private messages show the virologists expressed immediate skepticism about the quality of this data.

“They release these vague ‘environmental sampling’ results. What does that mean?” Holmes said on February 3.

Holmes, a visiting professor at the China CDC, had a contact there who had shared the unpublished environmental samples by February 9, 2020. The virus detected around the wet market was identical to the virus detected in patients, suggesting that humans and not an intermediate host had spread the virus around the wet market.

“I have seen the ‘environmental’ sequences (I hope this is OK to mention it Eddie?) – they are identical to the Wuhan backbone. But who knows what they are,” Rambaut said.

“If by identical you mean 100% like a lot of the SARS-CoV-2 sequences, my first guess would be it probably means they did not come directly from any animal,” Garry replied. 

“The environmental [sequences] are spectacularly uninformative,” Holmes said. “Pretty shocking if this is the best they have.”

“Yup. The environmental samples were entirely uninformative – I’m not actually convinced they’re environmental,” Andersen said.

“Kinda bummed that the ‘environmental’ samples didn’t show anything at all,” Andersen said in another message.

A 2023 analysis cowritten by Holmes, Andersen, Garry and Rambaut that examined the DNA of possible intermediate hosts in these very swabs, highlighting raccoon dog DNA in particular. Once again, their analysis received global press attention. The media likened raccoon dog DNA in these samples to fingerprints found at the scene of a crime. However, a more in-depth analysis of these samples has undermined this claim, showing no correlation between viral RNA and the DNA of a plausible intermediate host. One sample containing raccoon dog DNA highlighted by the virologists in particular was found to only contain a single read of the virus.

The “Proximal Origin” virologists’ private messages align more closely to the China CDC’s own interpretation of the samples it swabbed around the wet market. The Chinese scientists have argued the virus detected in the swabs was nearly identical to that found in humans, and emphasized that they took samples in the areas of the market where animals were sold. As a result, any clustering of positive samples with animal stalls was not meaningful.

After “Proximal Origin” was published, the virologists privately expressed other justifications for ruling out a lab origin — ones too shallow to be featured publicly in Nature Medicine.

“I’ve not heard of any cover-ups etc. [China CDC Director] George Gao has led most of the sampling and genomic work and he’s too dumb to set up a sophisticated theory,” Holmes said. “Why would the [Wuhan Institute of Virology] group then publish RaTG13 that would only help point the finger at them? Makes no sense.”

Lingering doubts

Two of the five “Proximal Origin” authors, Andersen and Lipkin, privately expressed that “Proximal Origin” had not truly ruled out all laboratory-based scenarios after their letter was published in Nature Medicine.

By April 3, 17 days after “Proximal Origin” first appeared in Nature Medicine, Andersen began “second guessing” in light of other scholarship. A 2013 paper posted to the virology blog Virological showed that serial passage experiments on coronaviruses tended to amplify a 12 nucleotide insert in the spike protein — like the furin cleavage site. These inserts were “arginine rich,” just like the furin cleavage site in SARS-CoV-2, Andersen said.

“The whole furin site being messed with in [tissue culture] has me second-guessing myself,” Andersen said on Slack. “When we went through this whole process, remember we talked about ‘passage might make viruses acquire these sites’? We couldn’t find a reference, but somebody just posted on Virological, which led me to this [paper].”

Garry described the pattern as “freaky.”

“I don’t think any of this new knowledge goes against what we said in the paper, but it does make our ‘definitely not passage’ argument weaker,” Andersen said.

Soon after, the Washington Post reported on State Department cables detailing a shortage of trained professionals at the Wuhan Institute of Virology. 

The news spooked Andersen. 

On April 17, just as Fauci cited “Proximal Origin” as evidence the virus was consistent with natural evolution from the White House press briefing room, Andersen fretted privately to his coauthors on Slack. 

“I’m a little worried about these ‘cables,’” Andersen said. “Because is it possible that they might have something? I’m putting all this to typical Trump BS smoke and mirrors (and just plain idiocy), but I’m not quite willing to die on this hill.”

More digging into the scientific literature showed the virologists had likely misinterpreted the meaning of the virus’ O-linked glycans, carbohydrate chains that are usually attached on proteins close to prolines, and were predicted to exist in SARS-CoV-2 because of the proline (P) in the furin cleavage site (PRRA). Because these carbohydrate chains can help shield viruses from a host’s immune system, the virologists reasoned that SARS-CoV-2 must have evolved naturally in the presence of an immune system. 

“The generation of the predicted O-linked glycans is also unlikely to have occurred due to cell-culture passage, as such features suggest the involvement of an immune system,” Proximal Origin reads.

But by a month after its publication, Andersen privately conceded that a 2017 paper showed this was likely wrong, and that this error toppled a key argument against the possibility of the virus arising out of certain lab experiments. 

“I really really want to go out there guns swinging saying ‘don’t be such an idiot believing these dumb theories – the president is deflecting from the real problems’ but I’m worried that we can’t fully disprove [cell] culture,” Andersen said. “Our argument was mostly based on the presence of the O-linked glycans – but they could likely play a different role.”

Andersen also reiterated a sticking point that each of the virologists had acknowledged in the weeks prior but which was neglected in “Proximal Origin” — that methods in modern virology allow viruses to be easily engineered without leaving a trace.

“We also can’t fully rule out engineering (for basic research) – yes, no obvious signs of engineering anywhere, but that furin site could still have been inserted via Gibson Assembly (and clearly creating the reverse genetic system isn’t hard – the Germans managed to do exactly that for SARS-CoV-2 in less than a month),” Andersen said in the same message. 

Even the “half furin cleavage site” in the bat coronavirus RmYN02 did not explain how a furin cleavage site appeared in SARS-CoV-2, Andersen said. 

“RmYN02 has an arrangement around that site, but it’s not this type of insertion,” he wrote. “I agree with you that it’s evidence for ‘this all occurs naturally,’ but it still doesn’t put a nail in the coffin of [the lab] theory.”

The other virologists brushed off Andersen’s concerns.

“Point taken that there truly could be intercepted ‘cables,’ but of what? We already know that the Chinese went into deep cover-up mode for example by shutting down the market and destroying the ‘evidence.’” Garry said. “It’s possible WIV characterized a NCoV-19 isolate earlier than the first noted cases in Dec I suppose, but that doesn’t make WIV the proximal origin of the virus. It’s also possible that the Chinese knew about a new respiratory virus spreading before the fish market cases.”

Andersen backed off on the question of engineering the next day, but again pressed his coauthors on the possibility of other lab experiments.

“While our evidence against engineering is very (very!) strong, our evidence against culturing isn’t (the presence of O-linked glycans probably controls activity of the [furin cleavage site] and isn’t a mucin like domain like we describe),” Andersen continued. “Again, I’m pretty sure this is all smoke and mirrors, but I’d need to see those actual cables before I put my head on the chopping block.” 

Andersen found little support among his coauthors, whose positions against the lab leak theory had hardened since the publication of “Proximal Origin.”

“Let’s face it, unless there is a whistleblower from the WIV who is going to defect and live in the west under a new identity we are NEVER going to know happened in that lab. Never,” Holmes said.

“The market is just too coincidental to ignore,” Holmes replied. “I see no reason to invoke lab escape whatsoever.”

The virologists received global media attention for “Proximal Origin.” 

The article reached millions within days, becoming the most high impact paper across any scientific discipline, according to Altmetric. 

“That’s tenure secured right there,” Andersen said in March shortly after publication.

At the same time, the virologists began to field angry emails from people who believed the pandemic leaked from a lab.

As the lab leak and zoonotic tribes began to solidify, Andersen appears to have pushed any lingering doubts to the side.

“I have decided that I am going to die on this hill, so I’ll talk to a few reporters and try to beat some sense into them,” Andersen said, linking to an article about the Trump administration tasking intelligence agencies to look into the Wuhan Institute of Virology. 

Andersen was not the only coauthor of “Proximal Origin” who doubted the soundness of the paper after its publication. 

Lipkin — who was privately mocked by some of the other virologists on the paper — later criticized his coauthors for overstating the case for a wet market origin, the committee discovered.

While the other “Proximal Origin” virologists claimed near definitive evidence for a wet market outbreak in a February 2022 preprint, Lipkin critiqued the analysis as “unverifiable” in an email to a reporter.

“Our colleagues fueled this with armchair epidemiology based on unverifiable data sets and terms like ‘dispositive evidence,’” he said. 

 “The Wuhan Institute of Virology has worked with bat samples and cultured bat viruses at BSL-2. This is a matter of published record – materials and methods in two papers.  This is unacceptable,” Lipkin said in a separate October 2022 email. 

‘Pango madness’

The virologists have pointed to the discovery of pangolin coronavirus sequences that resembled SARS-CoV-2 as critical to their 180 degree turn on the question of a lab origin.

“The features in SARS-CoV-2 initially suggested possible engineering were identified in related coronaviruses, meaning that features that initially looked unusual to us weren’t,” Andersen told the New York Times in June 2021. “For example, we looked at data from coronaviruses found in other species, such as bats and pangolins, which demonstrated that the features that first appeared unique to SARS-CoV-2 were in fact found in other, related viruses.”

But the new messages demonstrate these pangolin sequences were not in fact persuasive, especially as the virologists examined them more closely.

Holmes’ accounts differ, but either on January 30, two days before the teleconference, or February 2, one day after, University of Hong Kong virologist Tommy Lam contacted him about pangolin coronaviruses of interest.

By February 5, the virologists discussed the pangolin genomic data sampled from Guangdong and shared with Holmes by Lam. 

Though the pangolin coronavirus lacked a furin site and was too dissimilar from SARS-CoV-2 to be a progenitor, the receptor binding domain was a close match, containing six key residues that matched those of the novel coronavirus.

“Pango madness,” Holmes wrote. “Very similar to [the novel coronavirus] in [the receptor binding domain], sharing most of the key residues. Closer than RaTG13.”

Andersen expressed concerns about the “sketchy” quality of the data. Nonetheless, he renamed the Slack channel from “Project Wuhan Engineering” to “Project Wuhan Pangolin.” 

“With the pangolin virus possessing 6/6 key sites in the receptor binding domain, I am in favour of the natural evolution theory,” Holmes said in an email that included Farrar, Fauci and Collins on February 8.

South China Agricultural University announced on February 7 in a press release that it had isolated its own pangolin coronavirus from Guangdong reported to be 99 percent similar to the new coronavirus.

The virologists waited for the sequence to be released, hoping it would resolve their questions about the furin cleavage site and more closely align with the novel coronavirus than its closest known relative, the Wuhan Institute of Virology virus RaTG13.

In a February 8 email that included all of the virologists, Fauci, Farrar and Collins, Andersen said that he believed “everybody on this email” hoped the pangolins would alleviate concerns about a lab leak.

“The pangolin virus that was announced in the press conference might solve this issue if it has the furin cleavage site insertion which would be all but conclusive for the natural scenario,” Rambaut said.

If the genomes lived up to their hype and had the unique furin cleavage site, they would provide strong evidence that COVID-19 emerged from nature. Emails sent by Collins and Farrar suggest they were eager to see the sequences.

“Has the actual sequence of the pangolin coronavirus been released? That will be VERY interesting. Does it have the furin cleavage site?” Collins asked.

“I do not think so yet but…chasing!” Farrar replied.

However the virologists soon realized pangolins lacked the characteristics of an intermediate host.

Drafts of “Proximal Origin” acknowledged that “for the virus to acquire the furin cleavage site … this [intermediate] animal host would have to have a high population density – to allow the necessary natural selection to proceed efficiently – and an ACE2 gene that is similar to the human orthologue.”

Pangolin receptors appeared too dissimilar from human ones to facilitate the evolution of a virus so “preadapted” to spread in humans.

“I was trying to get a sense of how similar pangolin ACE2s were to human and whether replication in that host could lead to a receptor that’s quite finely tuned to the human receptor. Not very clear that that’s the case,” Andersen said on February 7.

In addition, closer inspection of the press release from the scientists in Guangdong showed the pangolin coronavirus was “up to 99 percent similar” to the novel coronavirus. The virologists acknowledged that it was unlikely to be a close enough relative to settle the virus’s origin.

“We really need to see if the pangolin data is as good as they claim,” Holmes said. “Indeed, it’s actually ‘up to 99%’ rather than ‘99%.’ That fooled me.”

Just days after telling the New York Times that the Wuhan Institute of Virology’s RatG13 was not a close relative of SARS-CoV-2, he appeared to privately grumble that the pangolin sequences were not close enough to dispel concerns about RatG13.

“Up to 99% is no good. There is a 342 bp stretch of RatG13 that is identical to CoV. Sigh,” Rambaut replied.

“The implications of a 99% similarity and a 99.8% similarity are pretty profound and at least would dramatically alter the discussion,” Garry said.

On February 18, Rambaut shared the sequence of what he believed was the “pango99” virus.

The virologists soon noticed that the pangolin coronaviruses did not hold the key to the pandemic’s origin.

“Pango99 (if that is what it is) doesn’t have the furin site,” Rambaut said.

“When they say ‘up to 99%” they mean an average of 90%,” Rambaut said on February 18.

“I think they have got over-excited with their results and claimed too much,” Holmes said. 

Garry said the pangolin receptor binding domain was unlikely to have contributed to a recombinant that led to the novel coronavirus. 

“No more of a smoking gun than RaTG13 as far as nCoV goes – not close enough to be the progenitor nor locally close enough to make a strong case that it might serve as a substrate for a recombinant that lead to nCoV,” he said.

The virologists updated their draft of “Proximal Origin” with the claim that finding a pangolin coronavirus with a similar receptor binding domain to the human optimized one in SARS-CoV-2 proved it was the result of natural selection. But privately the virologists were stuck on the lack of a furin cleavage site.

Far from dispelling their concerns about lab accidents, Andersen and Garry speculated about whether the pangolin coronavirus had been studied in a lab in serial passage and acquired the furin cleavage site that way.

“Without an insert, the closer it is the more likely the [serial] passaging theory becomes,” Rambaut said.

“Question is – did they recently realize that pangolins carry CoVs and then grew them in the lab to see if they could infect human cells? This is quite a high probability event,” Andersen said in a separate message. “Clearly none of these pangolin sequences were the source though.”

“The more pango sequences I see the less likely I find that they are intermediate – I think they’re just one of many animals with SARS-like CoVs,” he also said.

The virologists resolved to stay vague on the question of which animal could have served as the intermediate host.

“I don’t think we want to come down too heavily on the side of pangolins for now. I would just putting a bloody great question mark there. Or use a generic rodent sort of thing,” Holmes said on February 12.

Andersen may have remained concerned about the data quality, too.

“Four pangolin sequences just dropped as well – unfortunately these are similar to the previous and not similar in the [receptor binding domain],” Andersen said on February 17. “I’m starting to think that one pango that stands out might not actually be correct.”

A few days later, after the article’s submission to Nature Medicine, a peer reviewer opposed its publication because they believed the pangolin sequences were likely to resolve the origin mystery. The virologists pushed back, saying they were not a close enough match to have a direct relationship to the pandemic.

“The more sequences we see from pangolins (and we have been analyzing/discussing these very carefully) the more unlikely it seems that they’re intermediate hosts,” Andersen said. “They definitely harbor SARS-CoV-like viruses, no doubt, but it’s unlikely they have a direct connection to the COVID-19 epidemic.”

“Unfortunately none of this helps refute a lab origin and the possibility must be considered as a serious scientific theory (which is what we do) and not dismissed out of hand as another ‘conspiracy’ theory. We all really, really wish that we could do that (that’s how this got started), but unfortunately it’s just not possible given the data,” Andersen continued.

The pangolin data would not be published for weeks due to censorship in China.

“Unfortunately, they may not publish this any time soon because they have faced huge criticism in China, I think mainly from admitting that pangolin are illegally trafficked into China which apparently you are not meant to say. Very Chernobyl,” Holmes said.

When the pangolin sequences were finally published, only 1 gene was 99 percent similar to SARS-CoV-2.  The spike gene was only 84.5 percent similar, and had no furin site.

The pangolin coronaviruses were sourced from a single dataset first described in an October 2019 paper, outside inquiries later revealed. The data, renamed without attribution to the earlier paper, came from a handful of pangolins seized by Guangdong Customs in March 2019. 

The virologists’ private messages contrast sharply with their public depiction of the pangolin coronaviruses as a major revelation and turning point. 

Concerns about the Wuhan lab and EcoHealth Alliance

Though “Proximal Origin” only makes a passing reference to biosafety, the virologists expressed serious concerns about risky research and biosafety at the Wuhan Institute of Virology, the new messages show. 

The “Proximal Origin” paper does state that basic research involving the passage of coronaviruses has been done at biosafety level two “across the world.” 

Biosafety levels range from one to four (BSL-1 to BSL-4), with one being the most lax. 

But the virologists’ private messages show their concerns about the biosafety levels being hazardously low, specifically at the Wuhan Institute of Virology where scientists had a strong interest in conducting gain-of-function research that could make coronaviruses more dangerous. 

Andersen observed that potentially risky experimentation with coronaviruses in Wuhan was conducted at a “completely nuts” level of BSL-2, which is much less safe than the BSL-3 required for typical experiments with SARS and MERS and the BSL-4 level required for the most dangerous viruses. Biosafety level two would be insufficient for a virus as infectious as SARS-CoV-2.

“Performing these in BSL-3 (or less) is just completely nuts! IMO it has to be performed at BSL-4 with extra precautions,” Andersen said. 

Andersen also expressed concerns about gain-of-function research more generally.

“Honestly I’m not sure that type of knowledge is at all actionable, while, of course, being exceptionally dangerous. It only takes one mistake,” he said in the same message.

The virologists also recognized the lab’s “very strong focus” on spike proteins in coronaviruses — where the worrying furin cleavage site mutation resides.

“The main concern … is the kind of stuff that is being done – getting MERS-like viruses to infect humans, getting SARS-like viruses to cause disease in mice and infect humans, etc.” Andersen said. “There’s a very strong focus on the spike protein for all of that work.”

“Good idea to lay these arguments out for a limited discussion. And quite frankly so we can learn from it even if it wasn’t an escape – it easily could’ve been,” Rambaut replied. “If nothing else – the fact that we are discussing this shows how plausible it is.”

Because SARS-like viruses circulate in South China, the virologists also asked how a SARS-like coronavirus could have arrived hundreds of miles away in the metropolis of Wuhan without sparking other outbreaks along the way. 

Holmes remarked that all SARS-like coronaviruses had been found hundreds of miles from Wuhan in southern China. Holmes also noted that the Wuhan Institute of Viroogy found most of their coronaviruses in Yunnan Province.

“Interestingly, all the SARSr-CoVs that are capable of using human ACE2 were found in R. sinicus in Yunnan Province,” Andersen quoted from Shi Zhengli’s paper. 

“There are bat betaCoVs from Hubei [Wuhan’s province] but they fall into different clades and are not from R. affinis. The Wuhan group seem to sample almost exclusively in Yunnan. Must have loads in their freezers. So, [IN] that sense it’s no surprise that their virus is from Yunnan. BUT, if natural, what must mean is that there is a betaCoV from a bat from Hubei that is >96.5% similar to 2019-nCoV AND that there must be an intermediate host that is even closer still. Again, may all be natural. But I am struck by how differently this virus is behaving from SARS,” Holmes said.

Garry notes that the city of Wuhan was treated as an unlikely site of exposure to bat coronaviruses in an EcoHealth Alliance study.

“Sera from Wuhan – I kid you not – was used a control group,” Garry said. 

Andersen made a similar observation about RaTG13, the cousin virus identified at the Wuhan Institute of Virology.

“I believe RaTG13 is from Yunnan, which is about as far away from Wuhan as you can be and still be in China. What are the chances of finding a viruses that are 96% identical given that distance?” he said.

“Ebola got from Middle Africa to West Africa in 10-20 years,” Rambaut replied, another allusion to the controversial origins of the 2014 West African Ebola outbreak.

It’s clear others in the virology community were talking about the Wuhan Institute of Virology too. Holmes relayed a rumor from a professional conference that the lab had a novel SARS-like virus in the fall of 2019.

“I met this guy who said his mate at the Wuhan Institute of Virology had a human ‘SARS-like CoV’ sample from August 2019,” Holmes said.

More generally, the virologists demonstrated they were aware of the years-long debate within the scientific community in the years before COVID-19 emerged over whether new techniques in virology research could spark a pandemic. 

“The mind can play tricks and one of those tricks is denial,” Garry said. “SARS-CoV-1 escaped from Chinese labs 2, 3 or 6 times (depending on your source) AFTER the outbreak that killed 10% of people infected was over. Yes, Wuhan maybe getting too much of the attention – could be anywhere. We know two groups in Guangdong were doing metagenomics and growing CoV from pangolins perhaps for years. Escape via a custodian or researchers could happen from a lab and you would PROBABLY never know it.”

The virologists privately ridiculed the Wuhan Institute of Virology’s chief American partners — the nonprofit EcoHealth Alliance and Baric’s lab at the University of North Carolina — joking they had possibly released the novel coronavirus, even as they publicly supported these scientists. 

Andersen, Garry, Rambaut and Holmes had previously been critical of EcoHealth Alliance’s plans to expand its approach to pandemic prevention by seeking to collect every virus circulating in nature and predict which could spill over into the human species. 

But after the pandemic emerged from Wuhan, the virologists pivoted to defending EcoHealth’s research as essential to stopping the next pandemic.

Yet the virologists continued to make private gibes about EcoHealth, its virus collection program with the U.S. Agency for International Development called PREDICT, and Baric’s lab.

They mocked EcoHealth Alliance President Peter Daszak, as “Dastwat” and the “Grand Wizard of EgoHealth.” 

Garry proposed creating a diagram to outline the origin possibilities in play — which at that point included the virus escaping from the Wuhan lab after passage in cell culture.

“A diagram outlining the three scenarios with cartoons … Don’t make the cell culture passage scientist look asian (but maybe resemble an Ego guy),” Garry joked. 

“We need a cartoon picture of Peter Daszak to use in all the figures,” Rambaut agreed.

Garry followed up first with a cartoon of a mad scientist, then a sketch resembling Daszak.

The virologists noted that PREDICT underwrote EcoHealth’s work to uncover novel bat coronaviruses in Southern China, where COVID-19’s closest cousin viruses circulate. 

“One thing I find kinda funny here – all of this work getting bat samples was supported by PREDICT. So if they’re not able to predict the pandemics they themselves cause, then I’d say their program is in pretty bad shape,” Andersen said on February 3.

“Perhaps they had planned a press conference predicting which virus would cause the next pandemic but then it escaped from the lab early?” Rambaut said.

“Any [intermediate] animal being identified would be beneficial to them – otherwise we’re all going to point fingers at them telling people that they’re so shit that they can’t even predict the outbreaks of their own making…Too harsh?” Andersen said on February 7.

The virologists also expressed concerns about the work of Baric.

Drafts of “Proximal Origin” as early as February 3 state that “genetic data clearly shows that 2019-nCoV is not derived from any previously used virus backbone, including those recently posited by various conspiracy theories based on a 2015 paper in Nature Medicine,” with a reference to a coronavirus gain-of-function collaboration between UNC and the Wuhan lab that privately concerned the virologists.

“Proximal Origin” had ruled out a gain-of-function experiment gone awry, but privately the jokes continued.

In a February 10 message, Garry joked about the “next [novel coronavirus] diseases that escapes from anywhere – say a lab in North Carolina.”

Mere days after joking about PREDICT, Daszak and Baric as potentially culpable for the pandemic, the virologists received their first citation: A letter published in The Lancet circulated by Daszak condemned any questions about the Wuhan Institute of Virology as stoking prejudice and distrust, and cited an early version of “Proximal Origin” published on a virology blog.

By the time “Proximal Origin” published in the spring, the virologists appeared comfortable with their analysis supporting the continuation of the virus hunting and gain-of-function research they had privately expressed concerns about, even in light of concerns raised by other credible scientists like Ebright, Lipsitch, and University of Washington Professor Carl Bergstrom.

“Ebright/Lipsitch/Bergstrom are going to try to use this to end GOF research when I think this is going to be time we need it most,” Holmes said on April 17.

The ‘big ask’: The influence of powerful scientific institutions

The virologists glossed over many of the unresolved scientific complexities about the virus and the circumstantial evidence pointing to the lab in order to write “Proximal Origin” within a matter of weeks. 

The recently released messages depict a motive behind their rush to firmly conclude a zoonotic origin: Fauci and Collins, whose institution funded the research of Andersen, Garry and Lipkin, and Farrar, whose institution funded Holmes and Rambaut, appeared to push the virologists to publish something that would “dampen down” the lab leak theory.

As some of the most influential funders of scientific research in the world, Fauci, Farrar and Collins had their own motives for downplaying the possibility that gain-of-function research had ignited the emergent pandemic. Farrar said that a credible group of scientists should rein in the discussion about Wuhan’s labs “before that debate gets out of hand.”

Farrar passed along drafts and updates to Fauci and Collins. Farrar also sought their input as to whether it should be published at all. The virologists nicknamed Fauci and Collins the “Bethesda boys” after NIH’s headquarters of Bethesda, Maryland. 

Fauci and Collins simultaneously exchanged emails about whether the NIH could be linked to the Wuhan Institute of Virology’s research on RaTG13, the novel coronavirus’ close cousin virus. Fauci had been alerted by late January that his institute had funded coronavirus discovery and research in Wuhan through the EcoHealth Alliance. Collins also expressed concerns the theory could damage “international harmony.”

As the February 1 teleconference got underway — with Fauci, Farrar, Collins, the “Proximal Origin” authors, and the gain-of-function virologists including Fouchier and Koopmans on the line — the “Proximal Origin” authors exchanged thoughts on Slack. 

About 40 minutes into the call, the virologists exchanged enigmatic messages. 

“Big ask!” Holmes wrote. 

“Destroy the world based on sequence data. Yay or nay?” Andersen replied.

A few minutes later, Andersen makes another reference to an apparent request.

“Makes sense what he’s saying – but man, that’s hard to pull off,” he said.

The precise nature of the “big ask” remains unclear.

But it is clear that after the call, the group of virologists began accumulating evidence to suggest the virus was not engineered and shelving the evidence pointing to a lab.

“I think someone needs to lay out the science of this before it gets out of hand (and causes more formal investigations),” Rambaut said on February 8 on an email chain that included Farrar, Fauci and Collins.

Rambaut brainstormed about how to dig for stronger evidence to discredit the lab origin hypothesis than the idea that a scientist would only experiment on a known virus.

“I think we need stronger arguments than an assertion that no lab has done those experiments,” Rambaut said on February 8. “Is it possible to argue that A) a passaging experiment wouldn’t create the features we see? Or B) that there are logical reasons why someone wouldn’t do such an experiment?”

The virologists have insisted that they approached the problem as agnostic scientists, and intermittent Slacks do indicate some desire to weigh the lab origin possibility with fairness.

“I just think we need to lay out the features in the data and leave it open as to the cause,” Holmes said on February 3. “Just outline what needs to be explained and leave it like that. Irrespective of what the answer is, and will likely never know, these are really important bits of biology.”

“I think we should stick to our guns about the message and not tone it down just to get it published,” Holmes said on February 20.

But the new messages also show how that impartiality receded amid pressure from their powerful funders, their virologist peers, and editors from the journal Nature. 

The virologists had not even been grappling with the evidence for a week when Farrar asked if they could provide information to three of the most prestigious journals in the world.

Messages indicate that Farrar offered to use his position of power overseeing billions in scientific funding at the Wellcome Trust to place the piece in a high profile scientific journal.

“When can you update? Lancet, Nature, [New England Journal of Medicine] will all review immediately,” Farrar said on February 7. “Can I help with any of the editors?”

Farrar also said he would share a copy of their analysis with the WHO.

“Jeremy is pushing us to get a paper ready to go (prob. Nature),” Rambaut said on February 8.

“At this stage, it feels more like a school project than a full-on scientific paper to me, but we can definitely get it into shape,” Andersen replied.

Farrar made his aim clear in an email the same day.

“The theory of the origin of the has gathered considerable momentum not in social media, but increasingly among some scientists, in main stream media, and among politicians,” Farrar said. “The aim of this was to bring a neutral, respected, scientific group together to look at the data and in a neutral, considered way provide an opinion and we hoped to focus the discussion on the science, not on any conspiracy or other theory and to lay down a respected statement to frame whatever debate goes on — before that debate gets out of hand with potentially hugely damaging ramifications.”

“My preference is that a carefully considered piece of science, early in the public domain, will help mitigate more polarised debate,” Farrar continued. “If not, that debate will increasingly happen and science will be reacting to it. Not a good position to be in.”

Any ambiguity as to Farrar’s aim was dispelled in another email also dated February 6 in which he relayed a request from Koopmans — who the “Proximal Origin” virologists knew opposed addressing the lab leak theory at all — to “dampen down further the ‘conspiracy’ idea” as soon as possible.

“Possible to dampen down further the ‘conspiracy’ idea and make totally neutral? Talking with [Koopmans] last night and with the WHO meeting next week…both wondering whether actually publishing this sooner, but ruthlessly on the science….is worthwhile to put that flag down.” 

The virologists sought Farrar’s permission for every incremental question: whether to publish the paper, what preprint service to use, whether to add another author to the byline, even whether to circulate it to a wider email chain.

“Had a chat with Jeremy this morning,” Rambaut reported on February 12. “Really not much more to say. … Just that he still thinks it is important to get a matter-of-fact paper out there.”

“Yes, let’s just finish it,” Holmes replied.

Andersen would later tell a colleague that the process had been “a little rushed,” especially since he had been offline for several days, but cited “pressure from the higher ups to get it out.”

“Sorry the last bit had to be done without you … pressure from on high,” Holmes told Andersen. “All came together very quickly in the end. Jeremy Farrar and Francis Collins are very happy. Works for me,”

Both Farrar and Collins urged the group to publish the analysis as soon as possible, with particular attention on how it would be covered in the press.

“Keen to get out ASAP,” Farrar said.

“This is really well done, and I would argue ought to be made public ASAP,” Collins said.

On February 17, Farrar requested the draft be edited to tweak the sentence, “it is unlikely that SARS-CoV-2 emerged through laboratory manipulation of an existing SARS-related coronavirus,” to “it is improbable that SARS-CoV-2 emerged through laboratory.”

Farrar proposed holding a press conference. Discussions about whether to hold a press conference may have included Daszak, one email suggests. 

“It’s going to blow,” Holmes said on February 6. “Hence why Jeremy wants is thinking about putting something out. Hence the toned down version I just sent him.”

The group hoped the pangolin coronaviruses would prove a natural origin.

“Our main work over the last couple of weeks has been focused on trying to disprove any type of lab theory, but we are at a crossroad where the scientific evidence isn’t conclusive enough to say that we have high confidence in any of the three main theories considered,” Andersen said on February 8. “Like Eddie – and I believe Bob, Andrew, and everybody on this email as well – I am very hopeful that the viruses from pangolins will help provide the missing pieces. For now, giving the lab theory serious consideration has been highly effective at countering many of the circulating conspiracy theories.”

“We now have (and we will get more) pangolin data … We think we can tie this up even tighter with the next iteration and make a conclusive statement which will be the go to scientific statement to refer to,” Farrar echoed on February 8 — just a week after their analysis had begun.

Yet as the pangolin coronaviruses proved disappointing, Garry expressed concerns their data was insufficient.

“I have to admit likewise that I don’t really know the answers — maybe someone else does,” Garry said on February 10. “The data we have is just insufficient. … Even [the pangolin coronavirus] prob not helping (unless it magically has a furin cleavage site, which seems very doubtful).”

But under pressure to produce a published report in favor of a non-lab origin for the powerful funders, the group pressed forward with arguments and data they privately lacked confidence in.

When Andersen pitched the piece to the journal Nature, he said that it had been “prompted” by the higher ups.

“There has been a lot of speculation, fear mongering, and conspiracies put forward … Prompted by Jeremy Farrah, Tony Fauci, and Francis Collins … [we] have been working through much of the (primarily) genetic data,” he said.

Pressure from their peers, Nature

The virologists received pressure to distance themselves from the “conspiracy” position from colleagues in the field of virology who conduct gain-of-function research, namely Koopmans, Fouchier and Drosten. Their colleagues opposed giving oxygen to the lab leak theory at all, or even calling the public’s attention to the unusual furin cleavage site.

A dispute occurred between the “Proximal Origin” virologists and the gain-of-function virologists over whether ignoring the possibility of a lab leak or credibly considering and dismissing it would be more effective at quashing ideas they considered conspiratorial.

Specifically, acknowledging the possibility the virus emerged from serial passage could lead to headlines saying “top scientists consider this could have come from the lab,’” Andersen said.

But the virologists also knew a rejection of the lab leak theory without the appearance of having seriously considered it could lead to accusations of a “coverup.”

“This is already out there in full force so it’d be very important to discuss. Can’t just sweep that under the rug,” Andersen said on February 9.

While it ruled out lab origin scenarios, the preprint on the virology blog Virological released on February 17 included some the ambiguity the virologists were privately grappling with. Conclusively proving any origin theory would be difficult — perhaps impossible. 

“Although genomic evidence does not support the idea that SARS-CoV-2 is a laboratory construct, it is currently impossible to prove or disprove the other theories of its origin described here, and it is unclear whether future data will help resolve this issue,” the preprint read.

However, any discussion of a lab origin was scrubbed from the final draft of “Proximal Origin” after editors with the journal Nature questioned whether the piece would “feed or quash the conspiracy theories.”

The virologists apparently considered wielding Farrar’s influence to get around the rejection from the journal Nature.

“I get a sense that Nature might be a little gun shy … hence, we’d need to go all the way to the top,” Andersen said.

“Good idea – let Jeremy know and give him the rationale why Reviewer 2 was full of it,” Garry replied.

“The possibility [of a lab leak] must be considered as a serious scientific theory (which is what we do) and not dismissed out of hand as another ‘conspiracy’ theory,” Andersen told the journal. “We all really, really wish that we could do that (that’s how this got started), but unfortunately it’s just not possible given the data.”

In response to the journal’s pushback, the virologists made a tweak that would dramatically alter the meaning of “Proximal Origin,” changing their conclusion from saying that a lab origin was not “necessary” to stating that a lab origin was not “plausible.”

Andersen set to work on a press release titled: “The COVID-19 coronavirus epidemic has a natural origin, scientists say.”

The pressure to reject the lab leak theory within the scientific community was so immense that an anonymous whistleblower submitted a tip to a journalist in July 2020 that sought to expose the “Proximal Origin” virologists for having briefly entertained the lab leak theory. 

Holmes found himself defending the group for having ever considered the possibility. 

“I … strongly reject the idea that we should not have raised nor discussed the possibility of lab escape: as scientists we have to present all the data and discuss it openly. That’s all we did. To have not mentioned the possibility of lab escape would have been negligent,” Holmes said.

Media manipulation

The recent messages released by the committee show in new detail how the authors of “Proximal Origin” collaborated with Farrar, Fauci and Collins to curtail discussion of the Wuhan Institute of Virology in the media.

“The idea of engineering and bioweapon is definitely not going away and I’m still getting pinged by journalists. I have noticed some of them starting to ask more broadly about ‘lab escape’ and for now I have just ignored them – there might be a time where we need to tackle that more directly head on, but I’l let the likes of Jeremy and Tony figure out how to do that,” Andersen said on February 4.

“We’ll all likely be facing the lab escape or natural question head-on and should have a consistent response,” Garry said as they prepared for their manuscript to publish.

Several messages demonstrate the virologists’ desire to find evidence of the furin cleavage site and other aspects of the genome in nature in order to aid their “messaging.”

“Every little [bit] helps,” Andersen said on February 7.

“I’m actually rooting for some animal virus (bat, pangolin, something else hopefully not one of Kristian’s cats) to have a [furin cleavage] site,” Garry said.

“I think we’re ALL rooting for some animal virus here -would make the messaging so much easier!” Andersen said.

Across several messages, the virologists strategize like public relations professionals.

The earliest drafts of “Proximal Origin” begin with notes on features of the virus “already noticed or easy to discover.”

The group also expresses some anxiety about the press, and specifically the New York Times, finding out they are examining the origins of the pandemic at the request of Farrar, Fauci, and Collins.

Don McNeil, then a veteran science reporter at the New York Times, asked well-informed questions of Rambaut and Andersen after being tipped off by a source that the U.S. government was investigating whether the virus had any connection to the Wuhan Institute of Virology.

“Ignoring could further escalate, which I have to be very careful about. … one main problem I have too is that my name is on … ‘official’ things looking at this – so I need to be able to deflect potential future enquiries that could directly involve/name me,” Andersen said.

The virologists immediately began deflecting his questions.

For example, McNeil asked whether genetic manipulation could be detected in a virus.

“I gotta say, he pretty much nailed it. Let’s not tell him,” Andersen said. 

“Genetic manipulation leave signatures in a virus…No,” Garry said to the other virologists. “You could put the furin site in very cleanly.”

“Yes. But I didn’t say that,” Rambaut replied.

“I am thinking of just replying and saying that ‘I see nothing in the genome that would make me believe it has been genetically manipulated in the lab,’” Rambaut wrote. 

“Andrew’s response is credible and correct, but is not going to satisfy all the reporters,” Garry pointed out. “McNeil very credible [but] like every reporter can be mislead [sic].

Andersen also told McNeil that there was no way the virus could be a lab construct, repeating the false rationale at the center of “Proximal Origin” — that engineering is easily detected.

“A lot of conspiracy theories are talking about this being either a lab strain that had previously been produced or some new recombinant,” Andersen wrote to the reporter. “These rumors are demonstrably false – we would have been able to easily pick that up if that were the case, however it is not.”

Beyond ‘Proximal Origin’

While the messages clearly show the distortion of the issue in the media, the degree to which the authors of “Proximal Origin” influenced the U.S. government remains ambiguous. 

It’s not precisely clear to what degree the U.S. government was informed about concerns that SARS-CoV-2 was engineered and about “Proximal Origin” before its publication — even after the release of hundreds of pages of communications.

It is clear that Andersen briefed an interagency meeting organized by the State Department’s intelligence analysts soon after the paper published, an email obtained by U.S. Right to Know shows. Documents obtained by U.S. Right to Know shows that the article was incorporated by the Defense Intelligence Agency into its early COVID-19 origins report.

The Office of the Director of National Intelligence released a statement in April 2020 claiming that the intelligence community agreed with the “wide scientific consensus” that the virus had not been engineered, an apparent reference to “Proximal Origin,” but the genesis of this statement remains unclear. 

But there are conflicting accounts as to whether the U.S. intelligence community was alerted to the virologists’ concerns about a lab leak.

According to Holmes, Andersen informed the U.S. intelligence officials soon after flagging potential signs of engineering to Farrar and Holmes.

When Fauci first learned about the furin cleavage site, he said that if the virologists agreed there was cause for concern, that MI5 and the Federal Bureau of Investigation should be notified. Fauci, who was meeting with the White House and National Security Council regularly by that time, said that he would alert unnamed officials.

But Farrar’s memoir makes no mention of U.S. intelligence services. The memoir only states that Fauci alerted Collins, and that both were concerned about “anti-China rhetoric” from Trump.

“Patrick Vallance informed the intelligence agencies of the suspicions; Eddie did the same in Australia,” Farrar wrote. “Tony Fauci copied in Francis Collins, who heads the US National Institutes of Health (the National Institute of Allergy and Infectious Disease, which Tony heads, is part of the NIH). Tony and Francis understood the extreme sensitivity of what was being suggested, given the anti-China rhetoric coming from both President Trump and US Secretary of State Mike Pompeo.”

References to global intelligence agencies in the newly revealed messages provide little clarity. 

Lipkin — who said in a transcribed interview with the congressional committee that he gathered intelligence about outbreaks for the CIA — apparently told Holmes that “higher ups, including intel” were concerned about the furin cleavage site.

“If Lipkin says higher ups are concerned and intel involved it’s consistent with all we know too,” Garry replied.

In deflecting questions from the New York Times reporter McNeil, Andersen wrote to him: “National security? White House? Spooks? I wish my life was that exciting.”

But Andersen indicated in a sidebar to the other virologists that he was being dishonest: “I really … wished my life wasn’t this exciting…”

Twice — when contacted by the New York Times and by Lipkin — the virologists expressed little surprise at the news that the U.S. government and intelligence services were investigating the possibility of a lab origin.

Recent reporting states that Fauci cited “Proximal Origin” to persuade a CIA team to alter its assessment of the pandemic’s origin away from theories related to Wuhan’s coronavirus labs.

Despite all of the intrigue around “Proximal Origin” and the ongoing congressional investigation, the text of the article itself remains unchanged. 

In 2022, two of the coauthors were revealed to have undisclosed ties to the Wuhan Institute of Virology. Lipkin worked with the lab through a partnership between Columbia University and EcoHealth Alliance, an email obtained by U.S. Right to Know shows. And Holmes coauthored a paper about RaTG13, the cousin virus to SARS-CoV-2 identified at the Wuhan lab, the year before the pandemic emerged, an accidental release of sequences on an NIH database revealed. These connections remain undisclosed in the paper.

Springer Nature, the parent company of Nature Medicine, has yet to respond to calls to retract the paper, and the editor of the journal has blocked many of the article’s critics on social media.

While more than 5,000 people have signed a petition describing “Proximal Origin” as fraud, these signatures are equivalent to a small fraction of the more than 5 million people who have read “Proximal Origin,” and an even smaller fraction of the millions influenced by the accompanying media coverage that cast the lab leak theory as a far-fetched conspiracy theory.