Response of Dr. Kristian G. Andersen to questions relating to his email to Dr. Fauci recently released under FOIA

Nicholson Baker has shared the following response from Dr. Kristian G. Andersen with me. I am publishing it here until it is published somewhere else by Dr. Andersen himself. As of today, I have not seen it published anywhere.

From: Chris Emery
Subject: Re: request for clarification
Date: June 3, 2021 at 1:57:38 PM EDT
To: Nicholson Baker

Hi Nicholson,

Dr. Andersen received quite a few inquires about the emails, and many included similar questions. He put together the responses below to address them and help clarify the timeline and scientific process:

What data was available for our early conclusion?

For our preliminary studies there was very limited data with only about ten genomes from Wuhan and the genome of RaTG13, a SARS-related coronavirus found in bats, wasn’t yet available, nor were the coronavirus genomes from pangolins. We were well-aware of the decade-long work at the Wuhan Institute of Virology (WIV) researching SARS-related coronaviruses, including culturing of viruses in BSL-2 / BSL-3, and so gave serious consideration to laboratory escape as a possible origin for SARS-CoV-2.

What about the virus looked engineered to you then?

The features that stood out to us were the furin cleavage site (unique in the sarbecovirus sub-genus to which SARS-CoV-2 belongs, although it’s present in several betacoronaviruses, the genus of SARS-CoV-2), the SARS-CoV-2 receptor binding domain (unique at the time and our structural modeling suggested it might be a strong human ACE2 binder), a unique restriction enzyme site (BamHI) followed by a higher level of conservation towards the end of the spike protein, and a few other residues that had been observed to be important from research with SARS-CoV.

What did you mean by “inconsistent with expectations from evolutionary theory”?

This was our early conclusion based on a preliminary look at the data. As described above, there were features of the virus that appeared unique and at the time didn’t seem to have an obvious immediate evolutionary precursor. However, all the features in SARS-CoV-2 were identified in related coronaviruses in the first half of 2020, which largely invalidated the hypothesis of engineering and instead bolstered the argument for a natural origin. This is a textbook example of the scientific method, where a preliminary hypothesis is rejected in favor of a competing hypothesis as more data become available and analyses are completed. Today, based on significant additional data and analyses, we know that everything we observe in SARS-CoV-2 is highly consistent with evolutionary theory and natural emergence of the virus.

What made you change your mind?

In the days immediately following my email to Dr. Fauci, additional data was released (or we became aware of it), including the full genome of RaTG13, a coronavirus from bats, which presented many (but not all) key features seen in SARS-CoV-2. Most importantly though, following up on our preliminary analyses, we did much more extensive investigations — both on RaTG13 and other coronavirus genomes — to compare genomic diversity more broadly across coronaviruses. We looked carefully at all the literature from the Wuhan Institute of Virology (WIV), investigated common virus backbones and molecular, cellular, and cloning techniques used at the WIV / University of North Carolina, investigated sequencing data sets produced from WIV / EcoHealth, and performed kmer-based and recombination analyses on SARS-CoV-2. Many of these analyses were completed in a matter of days and allowed us to reject our preliminary hypothesis that SARS-CoV-2 might have been engineered.

By the time our peer-reviewed scientific paper “The Proximal Origin of SARS-CoV-2” was released, we had performed many additional analyses and discussed data, analyses and conclusions with many of our colleagues. Most importantly, significant additional data had become available, including related viruses from pangolins carrying a near-identical receptor binding domain, proving that all the features of SARS-CoV-2 exist in nature. Genomes from environmental samples at the Huanan market and more detailed reports about early cases had also been released. Combined, our work allowed us to conclude that SARS-CoV-2 is not a man-made virus and it is highly likely that it emerged naturally from a zoonotic reservoir / intermediate host, like so many other viruses before it.

It is important to understand that to this day, nothing has been proven and we are dealing with scientific uncertainty, based on currently available data. As we stated in our Proximal Origin article, we cannot prove that SARS-CoV-2 has a natural origin and we cannot prove that its emergence was not the result of a lab leak. However, while both scenarios are possible, they are not equally likely — precedence, data, and other evidence strongly favor natural emergence as a highly likely scientific theory for the emergence of SARS-CoV-2, while the lab leak remains a speculative incomplete hypothesis with no credible evidence.

What are the current theories around origins?

There are two main hypotheses — “natural emergence” and “lab leak.” When talking about a “lab leak”, it is important to consider that there are multiple versions of that hypothesis, all of which would have required the WIV to have been working on SARS-CoV-2 prior to the pandemic. There is no evidence to suggest they did, and in fact, available data suggest the opposite. Any version of this hypothesis also necessarily evokes a large-scale coverup, since the WIV, scientists, and Chinese authorities have denied having worked on the virus prior to the pandemic. The WIV also reported that staff tested negative for SARS-CoV-2 exposure, which would be inconsistent with a lab leak.

When we consider a lab leak, any theory must be able to explain several facts about SARS-CoV-2 and events surrounding its emergence, including both the chance of a lab leak itself (non-negligible, although to my knowledge it has never happened with an unknown novel virus), but also everything upstream (capture or engineering), as well events downstream (Huanan wet market being the early epicenter of the pandemic). There are three main versions to consider, (1) “capture-release”, which is a version of a lab leak where an individual collected SARS-CoV-2 from an animal host (e.g., a bat), brought it back to the lab, accidentally infected themselves, and the virus made its way to the Huanan market where it led to a superspreading event. A hypothesis of (2) “capture-manipulation-release”, which is similar to “capture-release”, but also involving scientists growing the virus in the lab (cells or animals), potentially via “gain-of-function” research. Finally, (3) “engineering-release” in which scientists created SARS-CoV-2 in the lab, accidentally infected themselves, and the virus caused a super spreading event at the Huanan market.

All of these scenarios, while possible, are highly unlikely. For the two “capture” scenarios, a researcher would have had to have found the next pandemic virus — SARS-CoV-2 — in its natural habitat, taken it back to the lab, accidentally infected themselves (or somebody else), and then started a super spreading event at a wet market (of all places). This is a very specific and highly unlikely series of events. All lab leaks of the past have been caused by known viruses, which have already been selected for their ability to infect humans (that’s how we discovered them) and grown to high titers in labs. A random sampling of novel viruses from bats, for example, are not selected for human infectivity. “Engineering-release” does not have to account for the likelihood of “capture”, but as described above, there is nothing to suggest that SARS-CoV-2 was engineered and it is highly unlikely that a scientist would be able to create something like SARS-CoV-2 in the lab from scratch.

In contrast, the scientific theory about the natural emergence of SARS-CoV-2 presents a far simpler and more likely scenario than the lab leak scenarios. The emergence of SARS-CoV-2 is very similar to that of SARS-CoV, with the same association with wet markets, timing around November, similar setting in a big Chinese city, and highly related viruses can be found all over China and more broadly across South-East Asia. The theory is also supported by all available evidence — which includes the early cases in Wuhan, clustering of early cases and excess deaths around the Huanan market (North of the Yangtze river — WIV is South of the river), environmental sampling of the virus from the market, and two lineages circulating at different markets early in the pandemic in Wuhan.

What should we do next

We should continue to investigate the origin of SARS-CoV-2 carefully, prioritizing more likely theories of emergence. I staunchly support further evidence-based studies and the most effective way to do this would be to continue to build on the work done by the WHO, while ensuring they have the necessary mandate and support. If our main goal is to truly understand what led to the emergence of SARS-CoV-2 to learn how to prevent and mitigate future outbreaks, such studies need to be collaborative and cannot consist of naive one-sided demands that are impossible to meet. The best place to start would be to continue from where the last WHO mission finished, but tracing back events at the Huanan seafood market, paying especially close attention to what animals were there, how they were sourced, and how supply chains may have been connected across the city.

Chris Emery
Senior Director of Communications
Scripps Research