Genetic models of the host-virus arms race

Interested individuals must follow Steps 1, 2 and 3 at this link on how to apply
http://www.ed.ac.uk/biology/prospective-students/postgraduate/pgr/how-to-apply

We expect evolutionary conflict to drive antagonistic co-evolution between hosts and viruses, so that like Alice’s Red Queen, both host and virus are “running as fast as they can, just to stand still”.

Consistent with an arms race, we often see adaptive protein evolution in viruses and in antiviral genes. Many people intuitively interpret this adaptive evolution as the outcome of a simple ‘tit-for-tat’ arms race, which reciprocally fixes beneficial mutations in the virus and host. However, although the intuitive model is appealing, it must be wrong—or at least too simple. RNA viruses and multicellular hosts cannot both be ‘running as fast as they can’, because the RNA viruses can ‘run’ (evolve) many thousands of times faster than the hosts.

This theoretical project will develop new and improved models (simulation models or analytical models) of the genetics of the coevolutionary process, as it occurs between rapidly evolving RNA viruses and their slowly evolving eukaryotic hosts. We expect that this work will be simulation-based initially, and will extend the classical gene-for-gene and matching-allele continuum models. However, we would be excited to discuss other approaches. The ultimate aim of this project is to test whether coevolutionary models that are motivated by the biology of hosts and viruses are consistent with the patterns we see in real-world population-genetic data.

This project will be supervised by Darren Obbard (IEB: evolutionary genomics, virus diversity, the evolution of invertebrate immunity; http://obbard.bio.ed.ac.uk/) and Luke McNally (evolutionary microbiologist, theoretical, comparative and experimental microbiology; https://lukemcnally.wordpress.com/). The successful applicant is likely to have a strong mathematical or computational background (physics, maths, computer science). Candidates with a strong bioinformatic background may also choose to include a larger component of data analysis, focussed less on model development and more on contrasting empirical data with models.