Integrating patterns across multiple genetic markers to infer spatial processes

Abstract

Landscape geneticists can take considerable advantage of differences in the action of evolutionary forces (mutation, migration, selection, and drift) on different loci within the genome. Appropriate comparisons among loci allow researchers to isolate and study the effects of these processes. For example, the organelles are typically inherited maternally (but occasionally paternally), and so will experience migration only when females or seeds disperse (males or pollen in the paternally inherited organelles). Thus, the comparison with biparentally inherited loci allows inferences about the differential migration rates of male vs. female animals or of seeds vs. pollen in plants. Conversely, all biparentally inherited nuclear loci should experience the same levels of migration and drift. Thus, loci that show unusually large levels of variation across the landscape (as compared with the bulk of loci) may be reflecting the action of spatially varying natural selection (local adaptation). Such comparisons are conceptually straightforward, but are complicated by the high intrinsic variability of stochastic neutral processes. Careful statistical analysis is needed to avoid over-interpreting differences among loci. Inferences will be most robust when both genetic and non-genetic data can be integrated, and the collaboration of landscape ecologists with geneticists promises to be particularly fruitful.