Spatiotemporal data and the prediction of bat virus spillover

Frameworks for predicting pathogen spillover have focused on phylogenetic patterns and ecological niche models, with less emphasis on mechanisms. A novel conceptual framework emphasizes the importance of linking hierarchical barriers to pathogen spillover across space and time. Because the initial barrier is the spatiotemporal distribution of shedding by reservoir hosts, predicting spillover requires understanding spatiotemporal infection dynamics in wildlife. We here analyze field studies of henipavirus and filoviruses in wild bats to quantify limitations in longitudinal sampling, biases in sampling effort, and relative importance of bat life history, study design, geography, and timing of sampling as global predictors of seroprevalence.