DFG FOR 1078 Natural Selection in Structured Populations

Breadcrumb Navigation


T1 - The genetic signature of phenotypic selection

Project leader

Joachim Hermisson
University of Vienna, Department for Mathematics

There is a striking difference in the view of the adaptive process in quantitative genetics and in molecular population genetics. In traditional models of quantitative genetics, phenotypic adaptation typically involves small shifts in the allele frequencies at many gene loci. In contrast, molecular models of “selective sweeps” usually assume that adaptation results from the sequential fixation of new beneficial mutations at only a few genes.

However, recent genome-wide association studies and genomic polymorphism data show that this view needs to be (partially) revised. As postulated by quantitative genetics, adaptation on most quantitative traits is indeed polygenic and the loci involved are often not independent. Models and statistical techniques to detect recent selective events thus need to account for polygenic effects, but their development is still in its infancy.


In the new project, we will develop methods to describe the genetic signature of polygenic phenotypic selection. Three objectives are devoted to

  • the development of a coalescent simulation tool to study polygenic selection;
  • a combined analytical and numerical description of polygenic footprints of adaptation under a wide range of selection scenarios; and
  • the development of statistical methods for the inference of selection and the application of these methods to DNA polymorphism data.