A central problem in evolutionary genetics is identifying genetic differences among populations of plants and animals that are biologically significant. For example, of all the genetic differences between humans and chimpanzees a few are biologically important and confer our ability to speak and walk upright while others have no discernable effect and are due to the random fixation of chance mutations since the divergence from a common ancestor. Likewise, there is tremendous interest in identifying mutations segregating in natural populations that are deleterious / disruptive of biological function. Since even for closely related species there are likely to be millions of genetic differences (most of which are likely non-significant), computational tools are needed for identifying those that are most important or interesting. The goal of this project is to develop statistical models for comparing patterns of genetic variation within and between species in order to identify which changes fall into adaptive (beneficial), neutral, or deleterious categories. The scope of the project is focused on development of mathematical models and statistical methods for fitting the models to the growing amount of genetic data. This project will focus on analyzing variation within populations of humans, Drosophila (fruit fly), and other ``model'' organisms where large data sets are being collected by other groups.
The broader significance of this work is the development of computer programs that can be used by other scientists to apply the methods/ models developed to their data. Doctoral and post-doctoral students will also be trained as a result of working on these research projects. Special emphasis will be placed on involving historically underrepresented minority graduate students in statistical genetic research. The scientific importance of this work is the potential discovery of meaningful genetic differences between species including humans and their closest relatives.
