Poliovirus epidemiology and immunology

We are the WHO collaborating institute on polio data analysis and modelling. We conduct epidemiological analyses and clinical trials of poliovirus vaccines to guide endgame strategy. Current work is focused on the vaccination response to vaccine-derived poliovirus emergence in Africa, persistent wild virus circulation in Pakistan and Afghanistan and development of next-generation sequencing methods for poliovirus surveillance.

Incidence of Circulating type 2 vaccine-derived polioviruses between January 2016 and June 2018 According to Time and Location. Full text article.

Infant immunity and the intestinal microbiome

We are interested in how the intestinal microbiome affects innate and adaptive immune responses at mucosal and systemic sites. A focus is infants in low-income settings exposed to multiple pathogens and their response to oral rotavirus and poliovirus vaccines. Currently, we are working with researchers in the UK, India and Zambia to characterise infant immune status and vaccine response in relation to the microbiome under steady state and perturbation (infection, antibiotics).

Infant immunity
Prevalence of intestinal pathogens at the time of vaccination with oral poliovirus vaccine by subsequent seroconversion status. Full text article.

Enterovirus evolution and epidemiology

Enterovirus serotypes are a significant cause of meningitis, respiratory disease and encephalitis (amongst others). They show diverse seasonal and multi-annual epidemic patterns that have not been adequately explained. We are using mathematical models fit to enterovirus time series from multiple countries to infer the key environmental and immunological drivers of disease incidence.

Dynamics of emergent Coxsackievirus-A6 (CV-A6) and Echovirus 18 (E18) causing human disease. Full text article.


We have recently developed integrated methods to analyse epidemiologic and genetic data in a Bayesian framework. Our current interests are in using these methods to estimate key infectious disease parameters, such as reporting rates and transmission heterogeneity, and in adapting these methods to next-generation sequencing data.

Inference of the reproduction number using 100 trees sampled from simulated sequences, compared with the estimates obtained using the true phylogeny. Full text article.