Presenter(s): Conner Lane
Faculty Mentor(s): Matt Streisfeld
Poster 58
Session: Sciences
Characterizing patterns of the genome for species spanning a large geographic space may influence the way we understand evolutionary differences between populations. For example, areas of the genome that are highly differentiated and are presumed to result from natural selection may instead arise from genome properties such as low diversity that exaggerate the differences between populations. As a first step towards making conclusions about natural selection in Mimulus aurantiacus, a phenotypically diverse plant occurring continuously across Southern California, we tested for correlations between genome statistics to characterize ways in which properties of the genome may be contributing to differences between populations. As an example, we expect high within-population diversity (π) to mask “real” differences between populations, leading to a lower differentiation score (FST), which we did observe. We also expected this negative relationship between π and FST to grow stronger with higher distance corresponding to more “real” differences to be masked by the within-population variation, which we only observed weak patterns of. This trend of correlations existing but only being weakly associated with geographic distance held true for all statistical comparisons. We conclude that M. aurantiacus does display specific genome properties that may act as confounding variables for processes we wish to examine such as natural selection. However, these patterns do not strongly correlate with greater isolation due to distance, indicating that genome properties may have less of an effect on distantly related populations than we might have otherwise assumed.