research

Bayesian phylodynamics

As pathogens spread through a host population, they evolve. If their rate of evolution is sufficiently fast compared to the duration of infection, then the genomes of pathogens sampled from a host population contain information about the process of disease transmission. In the group, we are interested in how we can best leverage this information to study rates of viral spread, and to examine the geospatial dispersal patterns of viruses, particularly from a Bayesian perpective. Technically, in the strict sense, only the former of these is “phylodynamics,” while the latter is “phylogeography.” We look forward to a future where there is less divide between these models in practice.

Reliable and trustworthy Bayesian phylogenetics

Underlying Bayesian phylodynamics is Bayesian phylogenetics. Bayesian inference is a powerful statistical tool for hierarchical models, a useful framework for incorporating prior understanding into models, and a convenient framework for jointly accounting for and propagating uncertainty. Phylogenetic models, however, can get rather complex, have to make (often stringent) simplifying assumptions, and the phylogeny itself is often an obstacle to inference. Thus, we are interested in ways that we can quantify the effects of our simplifying assumptions and double check that our analysis worked.