Single nucleotide polymorphisms (SNPs) represent an abundant and useful source of genetic markers to understand complex diseases. SNPs in coding regions (cSNPs) of biologically important genes are likely to functionally alter the protein product. We have used pooled DNA sequencing to evaluate the frequency of over 100 polymorphisms in cancer-related genes in Caucasian, African American and Asian populations. This allows the identification of SNPs useful in typing patient cohorts. To perform high throughput genotyping of cancer gene SNPs we have established 5' nuclease (TaqMan) assays for known variants in drug metabolizing enzymes, hormone receptor genes, immune response genes and chemokines and their receptors. These variants have been genotyped on cohorts of African American breast cancer, Caucasian lung cancer and Chinese lung cancer patients. We have developed a cocktail of polymerase and reference dyes for use in genotyping using the 5' nuclease (TaqMan) SNP genoptyping system. This cocktail provides better allele discrimination and is less expensive than the commercial product. We are adapting this method to 384-well plates that will allow a further reduction in cost, increase in throughput and reduced consumption of DNA. For assays that cannot be typed by TaqMan we have established a melt curve-based method that requires only low cost primers. Polymorphisms in several chemokine and chemokine receptor genes have been analyzed in an African American breast cancer cohort. Several significant associations were found that are being analyzed in larger cohorts to attempt to replicate the findings. Using high performance computing on the Cray supercomputer, an exhaustive search for short tandem repeats has been performed in the human genome. This data is in a browser and is searchable by gene or location. Thirty new genes that contain trinucleotide repeats in the coding region were identified that are candidate disease genes.
Showing the most recent 10 out of 40 publications