We use a variety of approaches from population genetics in order to study evolutionary processes. This work can be summarized in two general areas:
Natural population analysis – A major focus of the group is using population genetic theory to describe patterns of polymorphism associated with beneficial fixations, and the associated development of statistical approaches to identify these patterns in genomic data. We develop both maximum likelihood and approximate Bayesian (ABC) based methodology for identifying positively selected mutations and for co-estimating models of selection and demography (illustrating this difficulty are the three highly similar realizations of the site frequency spectrum under models of neutral equilibrium, background selection, and population growth shown here). Recent lab publications in this area include both those focused on methodology (e.g., Jensen 2014 and Ewing & Jensen 2016), as well as on empirical data analysis (e.g., Montano et al. 2015 and Renzette et al. 2016).
Experimental population analysis – As an alternative approach to characterize the evolutionary process, we also have a major interest in utilizing experimentally evolved populations as a way to control for many of the commonly confounding factors in natural population analysis. Relatedly, we develop statistical approaches to infer the distribution of fitness effects (DFE) and the underlying fitness landscapes from such data (with the beneficial tail of the DFE shown here in the A) absence and B) presence of oseltamivir treatment in experimental populations of influenza virus). Recent lab publications in this area include both those focused on directed mutagenesis (e.g., Bank et al. 2014 and Bank, Matuszewski, et al. 2016), as well as on mutation accumulation (e.g., Foll et al. 2014 and Bank et al. 2016).