Association Mapping in PD Linkage Regions

点击翻译按钮获取中文摘要

Linkage analysis is a powerful tool to identify the genes that lead to disease. Using this approach, we previously identified five regions, on chromosomes 5, 6, 8, 9 and 17 (Scott et al. 2001) that potentially contained PD risk genes. Recent results from independent genomic screens have now replicated our linkages to chromosomes 5 and 9. We have demonstrated the chromosome 6 region was due to Parkin gene mutations and follow-up on chromosome 17 indeed detected association between PD and the H1 haplotype associated with the tau gene. We now have significant association with a gene on chromosome 8 as well, supporting this linkage. However, linkage peaks in complex diseases present several new problems when attempting to use this method to identify the causal gene. First, the linkage peaks are very large, often 20-30 megabases (Mb) or more. This creates a great need for prioritizing candidate genes in a cost-effective and labor-saving approach, which led in part to our genomic convergence approach. Second, where the endpoint of a disorder like cystic fibrosis is a gene mutation, in complex diseases the endpoint of linkage analysis is identification era gene with "significant" disease association. But how significant? How does one know that a significantly associated gene is actually the gene producing the linkage result? Could a more important gene in the region, yet unexplored, be the actual gene that one is seeking? Also, is only one association contributing to these peaks? This project proposes to take a genomic approach to answer this question. We will apply a DNA pooling strategy (Hoogendoom et al. 1999) using single base-pair extension with denaturing high performance liquid chromatography to apply a new technique, "iterative association mapping" (IAM) to the entire linkage peak in an efficient manner. We will then focus on the strongest, most dominant peak of association, move from pooling to genotyping (Taqman) to identify haplotype-tagging SNPs on a small test set, then genotype the htSNPs on the full data set. Finally, we will test genes lying in the associated htSNP region for association with PD.