The individual-level approach: an original genetic field sampling scheme to optimize barrier detection in patchy populations

International audienceLandscape genetics focus on assessing the resistance of a landscape matrix by relating observed spatial genetic structure to landscape characteristics. When studying patchy populations, landscape genetic studies are usually based on population-level analyses, i.e. on estimations of genetic distances between pairwise populations. Because a low number of pairwise populations only provides a poor scan of landscape heterogeneity, we tested the capacity of an individual-based sampling at evaluating the impacts of landscape on genetic structures. Using both simulated and empirical genetic datasets, we assess the efficiency of an individual-level approach in detecting Isolation-by-Distance and Isolation-by-Barrier genetic patterns in patchy populations. Simulated datasets were created using CDPOP, a spatially explicit cost distance population genetics program: the individual-level approach proved to be a powerful methodological alternative to the more conventional population-level approach, especially in barrier detection. The empirical dataset concerned a widespread amphibian in north-eastern France, the alpine newt Ichthyosaura alpestris, in the outskirts of two transportation infrastructures (a highway exploited since 1969, and a high-speed railway, exploited since 1981). The study was designed so as to optimize barrier detection (30 to 40 year-old infrastructures, 417 genotypes sampled at the individual-level across 225 sites and use of 12 polymorphic microsatellite loci). However, no barrier effect could be detected, whatever the statistical analyses we carried out; on the contrary, we highlighted an unexpected spatial genetic pattern along the highway. Although it may not be possible to reject the hypothesis of an historical pattern, these results question the resistance of these transportation infrastructures on gene flow in this species: I. Alpestris may actually benefit from artificial aquatic networks (transverse and lengthwise water flow routing) for dispersal