The use of genetic markers to identify Quantitative Trait Loci / Bruk av genetiske markører til kartlegging av Quantitative Trait Loci
This thesis focuses on investigating some theoretical issues in linkage mapping and QTL fine mapping as well as the detection of QTL affecting clinical mastitis and somatic cell count in dairy cattle. Four papers are included in this work.
Paper I compares the power to detect heterogeneity of the recombination fraction using half-sib families with only informative offspring (OI) and half-sib families using all information (AI). Through simulation, the AI method was found to be more powerful than the OI method. The use of AI also led to unbiased estimation of recombination fraction. The Morton test was undertaken using both the AI and OI methods on the same simulated data sets for test for heterogeneity of the recombination fraction. Under the three simulated values of the recombination fraction (i.e. 0.05, 0.1 and 0.2) the use of AI resulted in the expected value of the statistic, whereas OI method yielded much larger values, leading to a much higher risk of false positive results. Therefore, caution must be taken when using the OI method to detect heterogeneity of the recombination fraction.
Paper II investigates three point linkage analysis carried out on the same simulated data set using both methods of OI and AI. Generally, the AI method gave unbiased estimates of the recombination fraction. The OI method resulted in biased estimates of the recombination fractions regardless of population size. The probability of obtaining the correct locus order was found to be more affected by the population size (i.e. number of meioses) than the choice of the method. It was concluded that although the use of the OI method results in the correct locus order with high probability, erroneous linkage map length may still be attained.
In paper III a genome-wide scan for quantitative trait loci (QTL) affecting clinical mastitis and somatic cell count (SSC) is presented. A genome-wise significant QTL affecting clinical mastitis was found on chromosome 6. Additionally four chromosome-wise significant QTL for clinical mastitis were located on chromosomes 3, 4, 14 and 27. A single chromosome-wise significant QTL affecting SCC was located on chromosome 8. No common QTL positions for the two traits were found in our study. This result indicates that the use of QTL affecting somatic cell count in marker assisted selection schemes aimed at reducing the incidence of mastitis should be carefully evaluated.
Paper IV investigates the robustness of a variance components QTL fine mapping method to unbalanced map densities and various sources of linkage disequilibrium such as selection, population admixture, genetic heterogeneity and genetic heterogeneity in recently mixed populations. The method assumes that linkage disequilibrium resulted from a mutation that occurred in the QTL a number of generations ago. Except for the scenario with genetic heterogeneity in recently mixed populations, results of the simulations show that the method was able to locate the QTL with only a small loss of accuracy. Therefore, the variance components method is appropriate for fine mapping QTL, also if linkage disequilibrium is in part due to other factors than the original mutation.
Key words: Heterogeneity of the recombination fraction, linkage analysis, mastitis, quantitative trait loci, variance components, fine mapping, linkage disequilibrium.
Pr. desember 2004: Ayman Sabry, Bioinformation and Biometrical Genetics Group Department of Genetics, Research Centre Foulum, P.O. Box 50, DK-8830 Tjele
