December 8 2010

12:00 LSB 2320

Andrew J. Eckert
Department of Ecology and Evolution & Center for Population Biology, UC Davis

Seeing the forest for the trees: Landscape genomics and the search for adaptive genetic variation in forest trees


Forest trees illustrate strong patterns of local adaptation. This is evidenced by the long history of provenance and common garden studies establishing the genetic basis and putatively adaptive nature of many quantitative phenotypes. The genes underlying adaptive traits in forest trees, however, have remained elusive. Multiple iterations of genotype-to-phenotype mapping (e.g. QTL and association mapping) have resulted in useful but limited knowledge about the underlying genetic architecture of functional polymorphisms. Missing from many QTL and association mapping studies, moreover, is a link to the environmental and ecological context of the detected genotype-phenotype associations. Landscape genomics offers a way to both identify and prioritize sets of genetic polymorphisms with respect to their role in local adaptation along specific environmental gradients. I will discuss two case studies, one for coastal Douglas fir (Pseudotsuga menziesii var. menziesii) and one for loblolly pine (Pinus taeda), in which landscape genomic approaches were taken to identify single nucleotide polymorphisms (SNPs) underlying responses to several climate gradients. In both cases, multiple SNPs were overly correlated to temperature and precipitation gradients after corrections for neutral patterns of population structure. These results will be discussed with respect to the feasibility of genome-wide discovery of adaptive genetic variation for forest trees, the implications of this discovery to resource management and the question as to which spatial scale is most relevant to local adaptation.















































































































































































































































































































































































































































































































this is idtest: