Quote of the Month

"Even apparently similar adaptations may be built from genetically different components."
-Theodosius Dobzhansky

Graph-theoretic genetic structure
Extant patterns of genetic diversity within natural populations of forest trees result from the interplay of several evolutionary processes, such as natural selection, genetic drift, and migration. When statistical methods developed to identify these processes are used in a hypothesis-testing framework, however, they do not often exhaustively divide the sample space into mutually incompatible outcomes. In other words, rejection of one hypothesis, a null hypothesis for example, does not necessarily imply that the alternative is true or even likely. The underlying reason is that several different processes can individually, as well as jointly, result in the same patterns (Nielsen 2005). For example, extended levels of linkage disequilibrium among genetic markers can result from physical linkage or any number of evolutionary processes such as natural selection, mutation, migration, and genetic drift. It is crucial to examine the sensitivity of the statistical methods I employ, therefore, to their propensity to yield false positive and false negative inferences. This is an active area of research in my laboratory, and I have examined this with respect to inferences of species level phylogenies in multiple publications (Eckert and Carstens 2008; DeGiorgio et al. 2014), as well as inferences from McDonald-Kreitman analyses (Eckert et al. 2013). Recently, I have also begun collaboration with Dr. Rodney Dyer to examine the performance of his Population Graph methodologies (Dyer and Nason 2004) to the identification of genetic polymorphism patterns that are consistent with those produced by natural selection (Zinck et al. in review). Central to these studies is the identification of the strengths, weaknesses, and limitations to the methodologies I use to make inferences about phylogeographic history and adaptation.

  • DeGiorgio, M., J. Syring, A. J. Eckert, A. I. Liston, R. Cronn, D. B. Neale, and N. A. Rosenberg. 2014. An empirical evaluation of species tree inference strategies using a multilocus data set from North American pines. BMC Evolutionary Biology. In Press.
  • Dyer, R. J. and J. D. Nason. 2004. Population Graphs: the graph theoretic shape of genetic structure. Molecular Ecology 13: 1713-1727.
  • Eckert, A. J. and B. C. Carstens. 2008. Does gene flow destroy phylogenetic signal? The performance of three methods for estimating a species phylogeny in the presence of gene flow. Molecular Phylogenetics and Evolution 49: 832-842.
  • Eckert, A. J., A. D. Bower, K. D. Jermstad, J. L. Wegrzyn, B. J. Knauss, J. V. Syring, and D. B. Neale. 2013. Multilocus analyses reveal little evidence for lineage wide adaptive evolution within major clades of soft pines (Pinus subgenus Strobus). Molecular Ecology 22: 5635-5650.
  • Nielsen, R. 2005. Molecular signatures of natural selection. Annual Review of Genetics 39: 197-218. 
  • Zinck, J. W. R., A. J. Eckert, and O. P. Rajora. The graph-theoretic properties of genetic structure among populations of eastern white pine (Pinus strobus, Pinaceae) and its relation to climate-mediated local adaptation. In Review.