Searching for Disease Genes Gets Easier

Continued from page 1

TR: Illumina added more SNPs to its newest chip, selected primarily from the African population. Why is this important?

DB: African Americans are more susceptible than Caucasians to some diseases, such as cardiovascular disease and Type 2 diabetes. To find the genetic variations responsible, we need to take into account the greater genetic diversity of the African population.

We had a chip with 550K [500,000] SNPs on an array, but that was not enough to capture the genetic variation in the African population. This population has high genetic variation because they have not gone through the evolutionary bottlenecks of other populations, such as emigrating from Africa or surviving the ice ages in Europe. So we added 100,000 SNPs from the Yoruba population [a West African population studied in the HapMap], which will be useful for studying the African-American population.

TR: Illumina recently acquired a gene-sequencing company, Solexa. Why?

DB: Once you find an SNP that is associated with a disease, you often need to sequence that region of the genome to find the causative variation. That means a lot of resequencing of specific parts of the genome. So we want to provide our customers with sequencing capabilities, and we think Solexa has the best technology.

TR: What else is in the works at Illumina?

DB: Our new product is iSelect, which allows people to design a chip with up to 60,000 SNPs of their own choosing. Such a chip could be used to diagnose a disease [in humans] or to study genes in other animals. For example, we have developed a bovine chip (see "Gene Chips for Cows").

TR: How would such chips work for diagnostics?

DB: We would need to narrow down the number of SNPs associated with a particular disease. Ultimately, the chips could be used to diagnose the disease and potentially to identify the best treatment for an individual. We're currently collaborating with DeCode, which has discovered SNPs associated with cardiovascular disease. We'll develop an assay for six to eight of those SNPs.