Island-Hopping with an E.I.D.

If you live in the U.S. and feel like Zika virus is getting closer to home, that’s because it is. Although there are no known cases of Zika transmission by natural vectors in the lower 48, experts have stressed that the virus’ spread remains unpredictable. The key word is ‘remains’ because, over the past year, Zika raced through large parts of South and Central America, leaving epidemiologists scrambling to both characterize and contain it. A conspiracy theory this is not.
In last week’s edition of Science, Faria and colleagues take an essential step in the process of describing the dynamics of this emerging infectious disease (E.I.D.). They trace the recent history of Zika virus using a dataset of new and previously reported genome sequences from across its range. Their paper also includes a large smattering of data points from Brazil, where the virus reached epidemic proportions (and surging notoriety among Western media) by mid-2015.

Correlation between airline passengers from Zika-infected countries arriving in Brazil per month versus number of suspected cases of Zika in French Polynesia on that same timescale (Figure 3C from Faria et al. 2016).


Zika virus is native to Africa and southern/southeastern Asia, and its island-hopping trek across the Pacific in an America-ward direction over the past decade has been impressive. As suspected given that spatial context, the new phylogenetic analysis of Faria et al. supports an Americas clade (mostly Caribbean and South American samples) and a Polynesian clade, as well as a sister relationship between the two. Divergence dating in BEAST suggests an origin of the Americas clade near the end of 2013, which is also consistent with extrapolations from Zika incidence data.
But reconstructing an exact trans-Pacific transmission scenario for Zika is trickier. Faria and collaborators try their hand at this using a couple of correlative approaches. First, they compile a database of air travel between Brazil and countries where Zika has been previously reported. This approach shows a >50% increase in such travel during the yearlong period bracketing the hypothesized arrival of Zika in Brazil (i.e., early 2013 – early 2014). Second, they compare posterior divergence date estimates for the Americas clade with dates for 3 Brazilian-hosted events of international significance: the 2014 World Cup, the 2014 Va’a canoe event, and the 2013 Confederations Cup soccer tournament. Each of the first two was previously advanced as the origin of Zika in the Americas; alternatively, only the Va’a event and the Confederations Cup tournament included competitors from Polynesia.
Philosophically speaking, inferring causal links between hand-picked historical events and estimated divergence dates (which can have wide confidence intervals) can be a slippery slope. Faria et al. (rightly) proceed with caution in this respect. While their datasets are most consistent with arrival of Zika virus in Brazil at the same time as the Confederations Cup, this is inconsistent with pre-Brazilian infection data (the Confederations Cup in Brazil was completed before the first Zika infection in French Polynesia was reported).
The biogeography of transmission for any modern, emerging infectious disease is likely to be complex AND inextricably linked to the activities and mobility of humans. By combining information on the latter with full genome datasets and results of phylogenetic and functional protein analyses, Faria et al.’s recent work is a step in the right direction.

This entry was posted in evolution, genomics, medicine, phylogenetics, Uncategorized. Bookmark the permalink.