Analysing gene flow in heterogeneous landscapes: why and how to use genetic graphs?

International audienceIn heterogeneous landscapes, when species occupy discrete habitat patches, ecological connectivity is influenced by populations’ topology. Graph-theoretic methods constitute a relevant tool to reveal this topology and better analyse gene flow. Despite growing interest in genetic graphs, a better understanding of when and how to use them is lacking.To fill this gap, we simulated gene flow between 50 populations in different landscape configurations and constructed genetic graphs using various genetic distances and pruning (link selection) methods. We then compared metrics derived from these graphs to analogous metrics describing the topology and connectivity of the dispersal network driving gene flow during the simulation.Genetic graphs consistently reflected the dispersal pattern. Pruning methods had more influence on the results than genetic distances, highlighting the crucial role of populations’ topology. Link selection based on maximal dispersal distances was more accurate to analyse ecological connectivity than when based on statistical inference of direct gene flow.Taking into account the pattern of population genetic structure is key to relevant analyses. Besides, graphs construction parameters should be closely tied to prior knowledge on species dispersal traits, when available. This study lays down a new framework to use genetic graphs in landscape genetics