In a few short weeks, Seattle-based biologist Trevor Bedford, 38, has emerged as one of the most famous epidemiologists in the world. His frequent tweets are seized upon by many of the globe’s top scientists and health policy makers. So far he has more than 170,000 Twitter followers, with thousands more joining every day.
But, unlike traditional epidemiologists, this disease detective working from his lab at the Fred Hutchinson Cancer Research Center, doesn't do field work to track down Covid-19 patients’ contacts. Instead, Bedford and a handful of colleagues — spanning the globe from Seattle to Basel, Switzerland, and Wanaka, New Zealand —analyse hundreds of virus genomes from patient samples to trace where outbreaks came from, how they spread from one corner of the Earth to the next and, most important, detecting early signs of infection clusters.
The team’s analytic approach relies on tracking how viruses mutate over time as they spread from person to person. In the case of the coronavirus, whose RNA consists of about 30,000 genetic bases or letters, it mutates about twice a month. These minor mutations tend not to change the potency of the virus. But they provide clues for genetic detectives to chart how they shift subtly over time, allowing them to create sprawling “family” trees, or phylogenies, that show how the coronavirus has spread from one part of the world or country to the next.
So far Bedford’s findings, which he summarizes promptly on Twitter, have been eerily on the mark, fueling his sudden celebrity status among fellow scientists and public health experts.
“Trevor Bedford offered some of the most careful analysis of this pandemic from the very beginning,” former Food and Drug Administration Commissioner Scott Gottlieb wrote in a March 14 tweet. “His estimates on the emerging epidemic in U.S. should be taken very seriously.”
Three weeks ago, when U.S. authorities still thought they might have the coronavirus somewhat under control, Bedford was among the first to argue that it had already been circulating undetected in the Seattle area for weeks. Virus-genome analyses suggested to Bedford that the very first patient in Washington in January, a 35-year-old man who had recently visited Wuhan, China, somehow infected someone else, allowing the disease to spread undetected for all that time around the Seattle area.
“There are some enormous implications here,” Bedford said in a nine-part Twitter thread on February 29 that has since been retweeted thousands of times. “I believe we're facing an already substantial outbreak in Washington State that was not detected until now due to narrow case definition requiring direct travel to China.”
This genome work differs markedly from traditional epidemiology that focuses heavily on identifying infected patients and tracking all their contacts. “Instead of talking to people about who they have been in contact with and shoe-leather epidemiology, we use the genetics of pathogens to see how they are spreading and how they are transmitting around the world,” says Emma Hodcroft, a molecular epidemiologist at the University of Basel who works closely with Bedford.
Genome sequencing has gradually become a more and more powerful tool over for tracking diseases. In the 2014 Ebola outbreak in West Africa, genome analyses helped trace the origin to a transmission strain that had been missed, allowing the disease to spread quietly for months in Sierra Leone. But that work took months to perform. Recently, genome sequencing has become a standard tool for tracing the source of bacteria-tainted produce.
Twitter has also become a crucial tool. Bedford says he has long written Twitter threads to accompany his scientific papers. But the coronavirus has moved so swiftly he hasn't had time for scientific papers lately. Once the first genome came out in January, “I basically started doing science over Twitter,” he says.