In June 2020, genetic sequences from 241 virus samples collected in Wuhan at the pandemic's start vanished from an online scientific database. Virologist Jesse Bloom from the University of Washington in Seattle recovered 13 of them by analyzing Google Cloud backups.
These sequences had disappeared from the NIH-maintained Sequence Read Archive (SRA). Dr. Bloom discovered their absence while reviewing a May 2020 PeerJ study spreadsheet that cataloged 241 SARS-CoV-2 sequences collected up to late March 2020 by Wuhan University researchers.
The spreadsheet showed the sequences were uploaded to SRA, but searches returned nothing.
Bloom's investigation revealed the 241 sequences were collected by Dr. Ai Shu Fu from Wuhan's Renmin Hospital.
Literature review led him to a March 2020 preprint by Fu, later published in Small, analyzing mutations in 45 nasal swabs from early Wuhan outpatients.
Bloom suspected these linked to the 241 sequences. Discovering SRA backups on Google Cloud, he successfully recovered 13 of the deleted sequences.
In his preprint (not yet peer-reviewed), Bloom argues there was "no scientific reason" for deletion, suggesting they were "removed to obscure their existence."
These sequences are vital for tracing SARS-CoV-2's origins. Researchers seek the earliest strains. Prior sequences from December 2019 Huanan market cases showed three mutations absent in later samples outside the market. Bloom's recovered sequences also lack these mutations.
As Bloom notes in the New York Times, this neither supports nor refutes a lab leak from a Wuhan P4 facility. Notably, these mutation-free viruses align more closely with horseshoe bat coronaviruses, suspected pandemic sources.
It bolsters evidence of circulation in Wuhan before official December 2019 reports, as market samples had already mutated. Exact collection details remain unknown but are essential for origin determination.
