The solution to many questions in evolutionary biology and agricultural science depend on the accurate estimation of the level of genetic variation and covariation present in the populations being studied. In this project, the investigator is developing software for estimating quantitative genetic parameters using statistical resampling methods, as well as theoretically verifying the applicability of these methods. In addition, novel techniques for comparing the pattern of genetic and phenotypic variation and covariation between two or more populations are also being developed and the software for conducting these tests provided. Tremendous advances in high performance computing over the last decade have made it possible to address many difficult statistical problems by repeatedly recalculating the statistical estimate after randomly resampling from the population or sample under study. Estimates of genetic variation within populations, which are essential for understanding how natural and artificial selection operate, have complex statistical properties, making them especially suitable for resampling approaches. The software developed in this project helps biologists in a variety of fields analyze their experiments in more powerful and innovative ways. This work is supported by the Computational Biology Activity and the Program in Population Biology (both BIO programs).
