Abstract
Much of the basic ecology of Ebolavirus remains unresolved despite accumulating disease outbreaks, viral strains and evidence of animal hosts. Because human Ebolavirus epidemics have been linked to contact with wild mammals other than bats, traits shared by species that have been infected by Ebolavirus and their phylogenetic distribution could suggest ecological mechanisms contributing to human Ebolavirus spillovers. We compiled data on Ebolavirus exposure in mammals and corresponding data on life-history traits, movement, and diet, and used boosted regression trees (BRT) to identify predictors of exposure and infection for 119 species (hereafter hosts). Mapping the phylogenetic distribution of presumptive Ebolavirus hosts reveals that they are scattered across several distinct mammal clades, but concentrated among Old World fruit bats, primates and artiodactyls. While sampling effort was the most important predictor, explaining nearly as much of the variation among hosts as traits, BRT models distinguished hosts from all other species with greater than 97% accuracy, and revealed probable Ebolavirus hosts as large-bodied, frugivorous, and with slow life histories. Provisionally, results suggest that some insectivorous bat genera, Old World monkeys and forest antelopes should receive priority in Ebolavirus survey efforts.This article is part of the theme issue ‘Dynamic and integrative approaches to understanding pathogen spillover’.
Highlights
Since the first human case was identified in 1976, Ebolavirus has caused recurring human and animal outbreaks in Central Africa, and a major human epidemic in West Africa
Research on possible phylogenetic patterns in Ebolavirus competency, and the degree to which hosts comprise a group of species whose feeding niche or other habitat requirements tend to bring them into contact with Ebolavirus reservoirs, has significant potential to advance our understanding of the ecology and epidemiology of the virus in the sylvatic setting
Number of individuals sampled for Ebolavirus RNA was the most important predictor and strongly positively related to host status, as was the number of individuals sampled in sero-surveys, the fourth most important predictor
Summary
Since the first human case was identified in 1976, Ebolavirus has caused recurring human and animal outbreaks in Central Africa, and a major human epidemic in West Africa. Human Ebolavirus transmission often arises from contact with infected wild mammals, which may serve to transmit infection [15]. Given that previous human epidemics have been attributed to handling infected carcasses [16] and to contact with bats [9,17], identifying shared traits and spatio-temporal patterns among sylvatic hosts may offer insights about the ecological mechanisms that drive spillover infection in humans. Research on possible phylogenetic patterns in Ebolavirus competency, and the degree to which hosts comprise a group of species whose feeding niche or other habitat requirements tend to bring them into contact with Ebolavirus reservoirs, has significant potential to advance our understanding of the ecology and epidemiology of the virus in the sylvatic setting. Using an automated approach to measuring variable importance, we predicted host status among 119 species and mapped the taxonomic distribution of presumptive Ebolavirus hosts in continental Africa
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
