Sampling strategy optimization to increase statistical power in landscape genomics: A simulation-based approach.

摘要: An increasing number of studies are using landscape genomics to investigate local adaptation in wild and domestic populations. Implementation this approach requires the sampling phase consider complexity environmental settings burden logistical constraints. These important aspects often underestimated literature dedicated strategies. In study, we computed simulated genomic data sets run against actual order trial experiments under distinct strategies differed by design (to enhance and/or geographical representativeness at study sites), locations sample sizes. We then evaluated how these elements affected statistical performances (power false discoveries) two antithetical demographic scenarios. Our results highlight importance selecting an appropriate size, which should be modified based on characteristics studied population. For species with limited dispersal, sizes above 200 units generally sufficient detect most adaptive signals, while random mating populations threshold increased 400 units. Furthermore, describe a that maximizes both sites show it systematically outperforms or regular schemes. Finally, although having more (between 40 50 sites) increase power reduce discovery rate, similar can achieved moderate (20 sites). Overall, provides valuable guidelines for optimizing experiments.