This project focuses on the development of new methodology for the statistical genetic analysis of quantitative traits. During the past year work has continued on the development and extensions of a method that can be used to accurately measure the heritability attributable to specific single locus effects (ROMP/ROOP). This method is being applied in several ongoing collaborative projects (see Genometric analysis of quantitative traits - applications. Work has also focused on the development of a new method that can be used to empirically estimate the type I error rate in genomic screening data without performing extensive simulations or theoretical approximations. This method uses the existing family structure and marker characteristics to obtain an appropriate estimate of the type I error rate of the genomic screen. In brief, a marker with a known chromosomal location can be used to generate a """"""""pseudo-trait"""""""" and tested for linkage to the remaining markers in the genomic screen that are non-syntenic. The distribution of the test statistic of the """"""""pseudo-trait"""""""" under the null hypothesis provides the basis for critical values derived as a percentage of this distribution that are then applied to a genomic screen between the actual trait and the markers used in the genomic screen. Other efforts include the development of a new method to measure the consistency of the estimates of heritability in longitudinal studies, evaluation of the variance components method under different ascertainment schemes, the use of single nucleotide polymorphisms (SNPs) as an alternative for micro-satellite markers in linkage analysis, and the use of moving averages and clustered markers as an alternative to multi-point linkage analysis.
Showing the most recent 10 out of 23 publications
