Development of statistical tools with applications in radiation hybrid and linkage mapping

In the last decade, many advances have been made in the area of radiation hybrid (RH) and linkage mapping. Radiation hybrid mapping (Goss & Harris, 1975; Cox et al., 1990) is a somatic cell technique for estimating the relative positions of markers along chromosomes. Because genomic sequences are unavailable in many agriculturally important species, there is still a demand for radiation hybrid mapping tools. Graph-based diagnostic tools are presented which allow visualization of the overall map quality and identification of error prone regions within a candidate RH map. Interval estimates for the locations of non-framework markers with respect to a selected “anchor” marker within a given framework map also are provided.; Linkage mapping is used to map quantitative trait loci (QTL). A screening method is presented for the special case where the trait of interest is categorical and a deterministic genetic model can be assumed. Our method is designed for the analysis of data collected from experimental outbred populations with minimal pedigree information, minimal marker map information and parental lines with unknown QTL genotypes (i.e., parental lines that may not be fixed). Under these conditions, simulation studies show that our likelihood based test for linkage provides adequate power even when applied to small experimental data-sets. Coverage of lod-based support intervals is assessed via simulation, and a diagnostic for accuracy of true coverage is presented. An efficient algorithm for exact maximization of the likelihood is described. Application of the method is illustrated using a real data-set.