Development of Linkage and Association Methods to Map Disease Genes

Abstract

Identification of disease susceptibility genes is one of the primary aims of contemporary genetic research. With the recent development in molecular biology techniques, large-scale gene mapping with a dense genome-spanning set of markers becomes a reality. The availability of markers throughout the genome has made linkage and association studies more feasible. In the first chapter, we review many linkage and association methods and point out the potential problems with current linkage and association analysis. In the second chapter, we modify two identity-by-state (IBS) test statistics of Lange (Lange K. 1986a, A test statistic for the affected-sib-set method. Annals of Human Genetics 50, 283--290; Lange K. 1986b, The affected sib-pair method using identity by descent relations. American Journal of Human Genetics 39, 148--150.) to allow for inbreeding in the population. We evaluate the power and false positive rates of the modified tests under three disease models using simulated data. When the population inbreeding coefficient is large, both the false positive rates and power are reduced when the modified test statistics were applied, although power remained high under a recessive disease model. Allowing for inbreeding is therefore appropriate at least for diseases known to be recessive. In the third chapter, we compute the proportions of affected sib pairs sharing 0, 1 and 2 marker alleles identity-by-decent (IBD) in an inbred population and express them in terms of higher order decent measures. We perform two consistency checks on the identity state probabilities and the two consistency checks verify our calculations. We did the same thing for affected sib pairs from first cousin marriage in an inbred population. In the fourth chapter, we study linkage and linkage disequilibrium (LD) simultaneously for single QTL using family data in an attempt to increase mapping resolution and reduce false positive rates. We estimate QTL allele frequencies, LD and recombination factions between the marker loci and the QTL locus and the QTL model parameters using an EM algorithm. After performing single analysis, we extend our model to study two marker loci simultaneously so that we can increase the accuracy of the estimations. Our simulation results show that our EM algorithm can give consistent estimates of all the parameters considered.