Dr Peter Foster

I am interested in model-based methods of phylogenetic analysis --- statistical approaches to making the family tree of life --- using DNA and protein sequences.

I have developed models of evolution that can accommodate changes in the process of evolution over time.

Together with my colleagues Martin Embley, Cymon Cox, Tom Williams, and others, we have tested the iconic Woese three-domains tree of life using these models, which showed the surprising result that if these better models of evolution are used on the same data more support is given to the eocyte hypothesis with an origin of the eukaryotes from within the Archaea.

This change in the overall shape of the tree of life has become accepted over the last decade.

I have written about these methods and the Tree of Life work mentioned above, some phylogenetic commentary, on eukaryotic microbial genomics and horizontal gene transfer, several genomics projects, and Anopheles systematics in collaboration with Anice Sallum.

I continue to develop a phylogenetics toolkit that implements these models, but also can be used for tree interrogation, and is a general phylogenomics toolkit that has been used in genomics pipelines to show the extent of horizontal gene transfer in eukaryotic microbes.

My work on supertrees with Mark Wilkinson at the NHM has led to a maximum likelihood implementation and to my knowledge the first Bayesian implementation of supertree construction.