Integrating spatial and genetic data
Understanding how pathogens spread is fundamental to developing effective tools to try and limit the public health burden they cause. However, this is often difficult to do. Imperfect surveillance systems coupled with spatial heterogeneity in health seeking behaviours mean that what we observe through surveillance may be a poor indicator of underlying patterns of disease risk. Pathogen genomes have the potential to help - as they allow us to infer the evolutionary relationship between pairs of pathogens which can then be compared to their respective locations. However, pathogen genomes themselves will also typically be sampled in a biased manner. We develop models that bring together different data sources to characterise pathogen spread, even when available genomes represent a small minority and biased subsample of all infections.
We work with a number of different pathogen types including bacterial and viral. In particular, we work with the national reference centers for Bordetella Pertussis and Listeria in France that have been systematically sequencing isolates from throughout the country and with partner countries throughout Europe. We study how the bacterium moves across different spatial scales and assess whether local vaccination patterns impact spread. We also work with questions related to the spread of arboviruses through the ERC funded ARBODYNAMIC project.