Using fossil records and simulations to develop spatially explicit demographic null models to detect selection from genomic data in forest trees

Detecting the genomic signature of selection has been a perpetual goal of evolutionary biology. The major difficulty of this task is that demography and selection can leave confounding genomic signals. Fossil records have long been used to corroborate population genetic analysis, however, inherent differences between fossil and genetic data have thus far impeded joint analysis. The objective of this research is to develop an interdisciplinary methodological framework to detect loci under selection in three European forest tree species. Coalescent simulations with a two-dimensional stepping-stone model forced with pollen-based population size fluctuations will be used to generate neutral expectations to detect loci under selection as indicated by unusual population differentiation across space. This approach could provide an exemplary case for using a biologically meaningful null hypothesis when detecting loci under selection, and could motivate similar research on other species.