Tracking a pandemic: Q&A with a COVID-19 detective

As COVID-19 spreads across the country and the globe, an international team of genetic gumshoes has been close behind.

These scientists track small changes in the virus’ genetic code as COVID-19 spreads from person to person. The changes act like fingerprints, helping researchers chart its global movements in near real time. 

Dr. Trevor Bedford, a computational biologist at Fred Hutchinson Cancer Research Center in Seattle, has followed the virus as it hopped from China to Washington to a cruise ship in California. He tracks the virus on Nextstrain, a public website he co-developed that provides informative and animated maps of viral evolution. (The genetic sequences are pulled from GISAID, an online repository where labs from all over the world post new genomic data.) Bedford is also publicly sharing his work on Twitter.

Nextstrain's phylogenetic charts — family trees for viruses — have helped guide the public health response in Washington state and beyond as the epidemic unfolds. We caught up with Bedford on March 12, as the U.S. ramped up efforts to combat the pandemic, to hear more about his detective work and find out answers to the questions he is getting asked most often.

How are you piecing together the transmission puzzle?

Viruses mutate very quickly and accumulate changes during the process of transmission from one infected individual to another. The analogy I like to give is the game of telephone. There are errors that occur as it gets passed on and can reveal who spoke to who.

These mutations in the genetic alphabet are generally really simple things. An "A" might change to a "T," or a "G" to a "C." This changes the genetic code of the virus, but it's hard for a single letter change to do much to make the virus behave differently. However, with advances in technology, it's become readily feasible to sequence the genome of the novel coronavirus. This works by taking a swab from someone's nose and extracting the RNA in the sample and then determining the "letters" of this RNA genome using chemistry and very powerful cameras. Each person's coronavirus infection will yield a sequence of 30,000 "A," "T," "G" or "C" letters. We can use these sequences to reconstruct which infection is connected to which infection.

What have you learned about how it has spread?

We can see all early samples from Wuhan are very genetically similar. Using the rate at which viruses evolve gives us a strong idea that this all came from a zoonotic event in late November to early December. After Wuhan, we see spread elsewhere in China. In February and March you can see areas like Italy and Washington state with clusters of viruses that are quite similar and suggest community transmission.

The best explanation in Washington state is that there was a traveler returning in January from Wuhan that sparked an outbreak. It was missed because the case definition was quite narrow. And then once testing criteria relaxed, we found a bunch of cases in the state. Another introduction from Wuhan is possible. The outbreak in Washington has gotten bigger and bigger, and we’re beginning to see some sparks fly off of that. The seeming sudden appearance of outbreaks across the US are not due to a sudden influx of cases. Instead, transmission chains have been percolating for 4-8 weeks now, and we're just now starting to see exponential growth pick up steam.

How fast is the virus mutating?

The rate looks to be about 24 mutations per year. Coming back to the game of telephone, that’s a mistake every second or third transmission. This rate of two mutations per month is similar to other RNA viruses like flu. This coronavirus has a longer genome than flu, so there are fewer mutations per base. None of the COVID-19 mutations look particularly interesting, but there a few things to watch for though. One is for mutations in the spike protein, which will be important for a vaccine.

What does the mutation rate mean? Is it cause for concern?

The mutation rate is as expected. Mutations happen randomly and are part of the lifecycle. Some mutations will break the virus. Other mutations can benefit it. We have been looking for this effect and so far there is not much evidence for adaptive evolution. That’s consistent with other outbreaks we’ve seen like Ebola.

The virus is moving rapidly. So is the science chasing it. How do you handle that?

This intersection of open science and a fast-moving outbreak is tricky to navigate. I absolutely believe that transparency is the best thing for global public health to be aiming for right now. Dialogue is the heart of science, and being able to have an open dialogue about hypotheses and results is a key component of open science.

I posted a hypothesis earlier this month about the connection between cases in Germany and Italy. Several colleagues whom I highly respect were skeptical of my conclusion, and their skepticism was warranted. I'm aiming for transparency in my thought process more than 100% correctness, and I try to keep up with the updates.

What happens next?

The cases seem to double every seven days, but it’s hard to predict what will happen as we move from containment to mitigation. We recently modeled projections for the burden of infections and deaths in two Washington state counties through April 7. The model shows that any social distancing that results in reduced transmission rates will slow the rate of growth of the epidemic, but only large changes in contact rate can interrupt ongoing transmission. We estimate that in the baseline scenario, on average across multiple simulations, there will have been roughly 25,000 people infected by April 7, but that this declines to roughly 9,700 total infections for a 25% reduction in contact, to 4,800 for a 50% reduction, and only 1,700 for a 75% reduction.

I've been hugely impressed by the thoughtful and deliberate actions taken to enact social distancing and combat the COVID-19 outbreak in Seattle and Washington state. People like Gov. Jay Inslee and Mayor Jenny Durkan have taken a science-based approach to their actions.

Additionally, capacity for testing is key to understanding the epidemic. If people can get results quickly, if they know if they need to isolate, we can reduce transmission. The hope is we could keep people out of the hospital. Widespread screening would have direct impact on transmission.

I believe the focus needs to be on testing and case-finding in the U.S. to slow transmission here. As viruses from elsewhere in the U.S. get sequenced, we'll learn about how connected these outbreaks are. Testing that is connected to epidemic surveillance can also guide additional mitigation efforts and social distancing policies throughout the country and the world.