On inferring and interpreting genetic population structure - applications to conservation, and the estimation of pairwise genetic relatedness

摘要: The presence of population structure is ubiquitous in most wild populations species. Detecting genetic and understanding its consequences for the evolutionary trajectories species has shaped a lot our process evolution. This delineation subdivision within plays an important role several allied fields, including conservation genetics, association studies, phylogeography, quantitative genetics. dissertation addresses methods to infer interpret subpopulation structure. In this regards, I discuss standing motivation developing new analytic tools, classic population genetics study imperiled freshwater turtle, Emys blandingii, development fast, likelihood based estimator structure, MULTICLUST, method pairwise relatedness structure. Our analyses midwestern blandingii detected considerable among sampled localities, revealed ancestral gene flow E. region north east from ancient refugium central Great Plains, concordant with post-glacial recolonization timescales. data further implied unexpected links between geographically disparate Nebraska Illinois. Our encourages decisions be mindful uniqueness across primary range. Analyses both simulated empirical suggests that MULTICLUST infers consistently (reproducible results), time effcient, compared popular Bayesian MCMC tool, STRUCTURE (Pritchard et al. (2000b)). also least bias, mean squared error estimating relatedness in full-sibling, half-sibling, parent-offspring, variety other related dyads, Anderson Weir (2007), Queller Goodnight (1989), Lynch Ritland (1999). Overall, lays grounds interesting biological statistical questions can addressed robust framework identification