Note: This is a freely accessible serialized version of Lab Leak Fever. Audio voiceover was AI generated for accessibility. Find an overview of all chapters here or consult the book website for further information. On March 4, 2023, while doing unrelated research, Dr. Florence Débarre randomly came across a new set of FASTQ files (a text file of nucleotide base sequences) on the GISAID database. Curious, she started investigating. Less than two weeks later, another viral information cascade would ignite the lab leak media universe ablaze again. This time, instead of the usual manufactured pseudo-events and trope-laden stories, a highly relevant scientific discovery supercharged its velocity and exploded in virality. A panic set in within the lab leak community; they were losing control over their viral narrative. Like Stuart Neil, Alex Crits-Christoph, Michael Worobey, and many others, Dr. Florence Débarre, a French theoretician in evolutionary biology, had started out being very open to lab-leak ideas, lauding the efforts and engaging regularly with DRASTIC, Alina Chan, Jesse Bloom, and other lab leak proponents for much of 2021 after the theory went mainstream. She wrote that it is “actually good scientific practice to explore different hypotheses” in response to criticism of Bloom et al.’s Science letter, the one that caused so much grief between Mike Worobey and Kristian Andersen. Florence, an extremely careful and meticulous researcher, had the habit of following up on certain lab leak ideas with investigative rigor. For example, she clarified with a web activity monitoring website that Shi Zhengli’s database of viral sequences “was not suspiciously taken down in September” to hide any sequences. It turns out the database first came online in April of the same year and never worked very well, dropping offline sporadically for months. It remained somewhat accessible until Feb. 2020, when hacking attempts finally stopped WIV researchers from putting it back online for fear of manipulation. Just as Shi Zhengli explained and Jane Qiu reported. Over the years, Florence has single-handedly cleared up about two dozen such falsehoods that the lab leak conspiracy myth cottage industry had made into the lore if my cursory counting is correct. She calls these fact-checks “niche threads” on Twitter, but they dismantled, debunked, and destroyed many lab leak talking points, like death by a thousand cuts. “The lab leak hypothesis survives in part because of poor fact-checking in the media,” she tweeted to explain why she followed up on all these circumstantial niche talking points. However, Florence did not have it out just for lab leakers; the zoonati would face the same type of scrutiny. Peter Daszak, for example, told me that Florence wrote to him countless times to fact-check statements he made in the past, asking whether he had supportive evidence and similar. He wasn’t alone; when Florence wants to get to the bottom of something, she becomes very tenacious and will not stop until she separates fact from fiction. For that, she has earned the respect of fellow debunkers and scientists and, unsurprisingly, has become another arch-villain to the lab leak community. This is because lab leak ideas and talking points tended to fall apart under her scrutiny, while points raised in favor of zoonotic spillover tended to hold up. Reality, on average, is easier to substantiate than made-up fiction. For her independent efforts, the female researcher has been not only severely harassed, smeared, and discredited like the rest of the zoonati she is now lumped in with but also encountered despicable misogynistic insults, stalking, and threats of violence. A high price for somebody who is an extremely private person by nature and avoids the spotlight. When Florence realized in early March 2023 that the new GSAID files she found were metagenomic data from environmental samples of the Huanan seafood market, which Chinese authors from Dr. George Gao’s CDC team had uploaded, she reached out to the Huanan market paper authors around Mike Worobey and Kristian Andersen. They, in turn, immediately started frantically downloading the data, which was about 500GB. “I was pretty convinced that we will probably never see these data,” Dr. Alex Crits-Christoph stated, describing his take on these hectic days and the drama that would follow. “But I have been thinking for over a year what we could do with it if they were ever published,” he laughed about how quickly he reacted. He was the fastest to look at the data. On the same day Florence had reached out, Alex downloaded the metagenomics sequencing data and pretty much worked through the night, looking first at the samples taken from the one corner of the western market where the wildlife stalls had been identified. “I found a full mitochondrial genome of a raccoon dog.” He remembered his excitement over the discovery. “And later that night, I remember bamboo rat and civet popped out as well.” Just as Mike Worobey and his market coauthors had predicted, wild mammals had been at the market, leaving their genetic footprint behind. So, what animals were possibly at the market in late 2019? “The first approach was the mapping to a few possible hosts, quick to see what is in there,” Alex recounted. They wanted to satisfy their curiosity. The next day, Prof. Eddie Holmes contacted a Chinese scientist, one of the few who tried to keep information channels open. To protect the scientist’s identity, let’s call the person Chen. Chen, whom Eddie described as someone “I trust completely” had previously told him that these metagenomics data had been messy, and they sequenced it multiple times. While not being part of the Chinese CDC, Chen had some insights into the sequencing data created for George Gao’s team. Eddie, opening his emails for me, explained how his first intuition was to talk to Chen and tell him that they saw the data uploaded on GISAID. Chen responded very quickly: Dear Eddie, Yes, George [Gao] has asked his colleagues to upload these data. In fact, some of these samples have been sequenced twice or even multiple times. [They] have uploaded all these data, including both SARS-CoV-2 positive and negative samples from the market. You can ask a group member to analyze them independently. I am happy to help if you have any questions. Best wishes, [Chen] That reply had been encouraging to Eddie; Chinese scientists were finally able to share some crucial market data from the preprint they published in 2022. A bit later, Eddie excitedly shared in another email to Chen that they had already found raccoon dog DNA in the environmental samples: As I’m sure you know, [the] most striking observation - which is of huge importance - is the high abundance of raccoon dog DNA/RNA. [...] So, we can now place susceptible animals exactly at the scene at the right time. He received no response from Chen after that. None of them did. Alex Crits-Christoph, Mike Worobey, and Kristian Andersen would all write to George Gao and his coauthors about the data, wanting to talk to them and open a collaboration. But radio silence. Unease set in with the international scientists. Suddenly, “and this was on like day three,” Alex described breathlessly, “the data disappeared from GISAID.” The Chinese authors, or somebody on their behalf, must have asked GISAID to pull it. “That really was pretty spooky, I think. We don’t know exactly what the heck was going through everybody’s mind.” Everybody tried again desperately to contact George Gao, William Liu, and others who had collected the data. They offered to work together on the analysis. However, the Chinese researchers were no longer willing—or able—to reply. Eddie, Kristian, and Mike thought something had to be done to get the data back online. They contacted WHO, which promptly engaged in a quiet meeting with the Chinese CDC about their meta-transcriptomics data. On 12 March 2023, some of us met with WHO and some members of SAGO (the WHO-convened Scientific Advisory Group for the Origins of Novel Pathogens) to discuss our observations. On 14 March 2023, the WHO convened a meeting with SAGO where some of us and representatives from CCDC presented our respective results. [...] This meeting constituted one of several efforts to establish a collaborative relationship with our colleagues at CCDC to share data and findings as rapidly as possible. We acknowledge that these circumstances are unusual. A summary by the authors would be made later. Meanwhile, led by Dr. Florence Débarre, all coauthors were busy analyzing the data and compiling a report for WHO. In another blow, they were all locked out of the GISAID database, which accused them of having breached the terms of use. A scary development. It was right around this time when journalists learned about the existence of these data and the upcoming report. Alex Crits-Christoph assumed that a member of SAGO leaked the information to journalists, but details remain murky. The press, in turn, started circling the scientists, smelling a scoop. Some asked for comments directly. The cat was out of the bag; many other press requests followed, and some of the scientists felt they had to give some statements. When the science writer Katherine Wu from The Atlantic reached out to Alex soon after, he tried to keep his statements general but explained candidly the rough outline of what happened. What they downloaded, what they analyzed, and that a raccoon dog was one of the species they found. He told Katherine Wu that she would be able to see more details in the report they are preparing, which was expected to come out soon. Katherine replied that she was going to publish her story before the report came out. “Oh my God,” Alex recalled. “The rest of the conversation was me telling her that this is a horrible idea.” But according to Alex, the reporter
