Some of the most valuable data for understanding how evolution works in natural populations is individual-based longitudinal data from pedigreed populations. Longitudinal data on individuals provides possibilities to link aspects of phenotype to life histories and fitness. Pedigree data allows inference of the genetic basis of variation in phenotypic traits, based on patterns of similarity of relatives. With collaborators at the University of Edinburgh and elsewhere, a portion of our research revolves around the study of the selection and genetics of a range of traits in Soay sheep from St Kilda (pictured) and other long-term animal datasets from around the world (see, for example, here and here).
Evolutionary genetic theory
We use analytical and computational approaches to understanding what patterns of genetic variation are expected in nature, and also of how to interpret observed patterns in microevolutionary parameters, including both aspects of genetics and selection. We have an ongoing interest in the patterns of genetic variation that are generated by complex landscape arrangements, especially in dendritic systems, which characterize all freshwater landscapes. Recent work has formalised the meaning of total vs. direct effects of traits on fitness in evolutionary quantitative genetic theory (see here). Ongoing work on a developmental evolutionary quantitative genetic theory seeks to develop a general theory of the evolution of the full joint distribution of phenotype (see some initial progress here).
Methods and software for empirical microevolutionary studies in nature
Analysis of data from natural populations is often very challenging. Datasets are often incomplete due to practical realities such as limited molecular information to resolve pedigrees, and/or imperfect detection of individuals for recording of life history information. We work on developing statistical tools to link fundamental evolutionary genetic theory to real data from the field. R packages include pedantics, and gsg (gsg faq).
Many of the lab’s empirical problems are addressed through collaborations. Some of these include analyses of data from long-running studies of Soay sheep with Josephine Pemberton, Loeske Kruuk, Dan Nussey (Edinburgh) and othersgreat tits with Ben Sheldon (Oxford) and others, , and of song sparrows (pictured) with Peter Arcese (University of British Columbia) and Jane Reid (Aberdeen).