The integration of disparate data about a biological system is critical for maximizing what we can learn from them. Typically, the information contained in one single data source is too weak to convey a sufficient evidence base for understanding the whole system. However, combined together these disparate sources can strengthen each other and provide fuller understanding of the system. While the statistical methodology and benefits of effective data integration are ubiquitous to almost all complex systems, this proposal focus on multi-host pathogens of humans and animals, and how best to enhance serological data - the most widely available source of information about the prevalence of infectious diseases. As a reflection of the immune system’s memory, serology is information rich. However, only when integrated with other information from the host and pathogen can serology be used to understand the epidemiological processes underlying disease prevalence. The overall aim of this proposal is to develop novel analytical tools to integrate data generated from serology surveillance with metadata from the host and pathogen, as well as sampling design and environment into mechanistic models that will enable a fuller understanding of the epidemiological process underlying them, reconstruction of the disease dynamics, and ultimately more effective strategies for disease control.

These tools will be developed and applied to three case-studies of public health, economic and conservation importance: i) Parvovirus in dogs and lions in the Serengeti ecosystem; ii) Bat influenza in vampire bats in Peru; and iii) Avian and equine influenza in horses in Mongolia.

First published: 12 September 2014

<< 2015