One goal of the HapMap project is to help reserachers find SNPs associated with human disease. Josephine Hoh at Yale University's School of Public Health and colleagues at Rockefeller University, New York, and the National Eye Institute in Bethesda, have identified an SNP associated with age-related macular degeneration (AMD), a major cause of blindness in people over 60. The SNP is in the gene for complement factor H, leading to a tyrosine to histidine mutation. The researchers studied 96 patients with AMD and 50 healthy controls, and measured the frequency of over 116,000 SNPs in each group. “For the initial screen, we used Affymetrix's set of 100K SNP chips,” says team member Robert Klein. “To identify the putative causal mutation, we used PCR to amplify each exon in a number of samples and then resequenced to find all variants in the exons.”

Josephine Hoh uses SNP arrays to find mutations associated with disease.

Hoh and her colleagues found that caucasian patients with AMD are at least four times more likely than usual to have this SNP. How the change causes AMD is not yet known, and one of the next directions for her lab “is to figure out the functional mechanism of complement factor H in the pathogenesis of AMD”, says Hoh. There are a few clues. The amino-acid change lies in a part of factor H that interacts with C-reactive protein and heparin, both known to be associated with AMD. And factor H is known to regulate components of the immune system that are found in drusen, fatty deposits that accumulate in the macula with age. In people with AMD, the drusen are larger and more numerous, killing cells needed to nourish adjacent retinal photoreceptors, which eventually results in loss of sight.

SNP mapping is also underway in animal models of human disease. Kent Hunter at the National Cancer Institute (NCI) in Bethesda, Maryland, uses SNPs to look for cancer-modifying genes in mice. “Ultimately, we hope to identify the particular polymorphic gene or genes that modulate metastatic efficiency,” he says. Maxwell Lee at the NCI is interested in how genetic variation determines gene expression and phenotypes in human cancer and uses SNPs to search for epigenetic markers. “We need to understand more dynamic aspects of the genome including interactions between SNPs and other downstream targets such as chromatin, DNA methylation and gene expression,” he says.

Caitlin Smith