NEW YORK (GenomeWeb) – An international team of researchers has used genomic data to reconstruct how the Ebola virus spread during the recent West African epidemic.
Researchers led by the University of Edinburgh's Andrew Rambaut analyzed the genomes of more than 1,600 Ebola viruses to develop a picture of how Ebola moved within and between Guinea, Liberia, and Sierra Leone during the 2013 to 2016 outbreak. Some 28,600 people were infected and 11,300 died during that time, according to the World Health Organization.
With the phylogenetic data they collected and analyzed, the researchers examined how factors like geography, climate, and demography affected the spread of Ebola. As they reported in Nature today, they found the virus typically spread according to a gravity model in which it tended to move between large population centers.
"If you use sequences ... because of the evolutionary rate of viruses like Ebola or like flu or like rabies, essentially mutations ends up encoding information about where it's been," first author Gytis Dudas from the Fred Hutchinson Cancer Research Center said. "With enough sequences, you can actually, with a fair degree of certainty, reconstruct the entire history of the virus."
Molecular clock dating traced the most recent common ancestor of the West African Ebola outbreak as arising between December 2013 and February 2014, while phylogenetic data indicates that this ancestor was in the Guéckédoc prefecture in Guinea. This suggested to the researchers that the epidemic began there in late 2013.
The virus then spread out of Guinea into Sierra Leone in early April 2014 and then into Liberia and back into Guinea, according to the phylogenetic tree they constructed that included data on when and where samples were collected. By the middle of September 2014, the researchers noted that 500 new Ebola cases were reported each week in Liberia, mostly due to an outbreak that included its capital, Monrovia. By December 2014, efforts to control the epidemic in Sierra Leone began to have an effect, and it began to come under control in Liberia and eastern Guinea in March 2015.
By linking their phylogenetic tree to geographic data on where the viral samples were obtained, the researchers found that five factors — including the size of both the source and the destination populations and geographical distance between population centers — influenced viral spread, according to Rambaut. They gleaned that the Ebola virus tends to move between regions that are geographically close — half of all virus dispersals were between locations less than 72 kilometers (45 miles) apart.
In particular, Rambaut and his colleagues reported that this dispersal followed a gravity-model dynamic in which the virus tended to spread from one large population center to another large population center, rather than among smaller spots.
To a certain extent, international borders served to block the spread of Ebola, the researchers found, even though the borders are rather porous. Rambaut noted, however, that transmissions still occurred between regions adjacent to international borders. He added that countries in the region began to close their borders during the epidemic in an attempt to limit the spread of the Ebola virus, but that those efforts were too late as the virus was already established within the countries. Dudas also said that after border closures, viral migration intensified within each country, as it decreased between countries.
Still, a few transmissions occurred outside these three countries, though an epidemic didn't take hold. Under their model, the researchers suggested that while these countries were at risk, their distance from areas of active transmission and their borders prevented an epidemic from becoming established there. Dudas likened it to a forest fire: even though there was tinder, the sparks didn't reach it to cause a fire.
With the advent of portable sequencing devices, Rambaut said that analysis like this could be conducted in the early stages of an outbreak as part of the response. He and his colleagues found that they could discern the same broad patterns that they reported in their paper from only the samples collected until October 2014, suggesting that such information could have been available early on to influence outbreak response.
Viral sequencing could "perhaps inform interventions at that time to change the trajectory, potentially, of the outbreak," Rambaut added, noting that it could inform contact tracing efforts as well as reveal connections between outbreak locations.