Genomic Tracing of Covid-19 Epidemic in Seattle

Among numerous news items related to the Covid-19 epidemic, I recently came across an interesting item which included some new and interesting information. I am referring to a post by Trevor Bedford, who is an Associate Member at Seattle’s Fred Hutchinson Center in the Vaccine and Infectious Disease Division. He is also an Affiliate Associate Professor in the Department of Genome Sciences and the Department of Epidemiology at the University of Washington.1

The post to which I am referring is a blog titled “Cryptic transmission of novel coronavirus revealed by genomic epidemiology.”2 Bedford and his team track the patterns of transmission and spread of pathogens by using the genetic mutations in the pathogen’s genes. This is very similar to the tracking of patterns of migration and ancestry in human history using the mutations in human genome, especially when we know the rate at which certain mutations on average take place.

In terms of the novel coronavirus, we know that the rate of mutation is roughly two per month. Moreover, we know that on average, it takes about 7 days for the infection to transmit from one person to the other. Combining these two pieces of information, the transmission of the virus can be visually represented as follows:

Here each mutation is represented by a change of color. Using this information, by sequencing the genome of the virus from two different infected patients, it is possible to tell if they are related (i.e., one is the parent of the other) and if yes approximately through how many transmissions or over what period of time.

This is exactly what Bedford has done on two infection samples from COVID-19 patients in Seattle. The first case, known as USA/WA1/2020, is the first identified case in US. The patient had returned from Wuhan to Snohomish County on Jan 15. The second case, detected around end of February, is known as USA/WA2/2020 and is a “community case,” which means the patient had not traveled to Wuhan and had not come into contact with a known Covid-19 patient. The remarkable result that Bedford and his team arrived at was that the WA2 case was identical to the WA1 case expect for 3 extra mutations. According to Bedford, based on the statistical mutation rate of the virus, this means that with overwhelming probability (~97%), WA2 is a descendent of WA1. Most probably, the infection has been transmitted undetected from WA1 around 5 times before getting to WA2. Bedford calls this “cryptic transmission.” One can conclude that in all likelihood there have been many cases of infection with mild or no symptoms at all. Bedford’s analysis shows that the number of infections in WA should be around 570 (as of March 2), with a 90% uncertainty interval of between 80 and 1500 infections.

What is remarkable about this conclusion is the number of unreported or unnoticed infections, most likely due to lack of severe symptoms. This underlines the importance of more testing to identify possible patterns of infection. For example, random checking of various individuals could provide a basis for a more accurate statistical picture of the spread of the infection. But I wonder if this also could be a sign of hope, in the sense that the death rate of the disease may be much lower than the nominal 2% often quoted. Moreover, I wonder if this means that a large number of people have developed (or will develop) immunity to the disease due to mild infection, and as a result, eventually the infection would dissipate because of herd immunity. Whatever the case, the picture is fast evolving and we must wait for new information.

~~ Moe Azadeh

References

https://bedford.io/team/trevor-bedford/
https://bedford.io/blog/ncov-cryptic-transmission/