October 28, 2002
Perlegen finds 1.7 million SNPs

Perlegen, a two year old spin-off of Affymetrix, has announced that it has compared the DNA sequences of 50 people and identified all the Single Nucleotide Polymorphisms (singe genetic letter differences) found in that group of people. Perlegen is keeping that information to itself to use business deals with pharmaceutical companies to identify which genetic variations cause adverse drug reactions:

Yet there was only a muted celebration a few weeks ago, when Perlegen's scientists decided they'd found the last of the 1,717,015 SNPs that biotech firms have been seeking since the human genome was sequenced in 2000.

"We cracked a single bottle of cheap champagne," quipped Perlegen chief scientist David Cox.

Two simple reasons explain Perlegen's restraint.

Although the company claims to have found just about every SNP in creation, the scientific community hasn't the proof. Perlegen won't publish its SNP map. Instead it will try to recoup its investment by helping drug firms use these subtle genetic variations to determine why some people react badly to medicines -- or get sick in the first place.

Perlegen will also use this data to look for genetic variations that predispose people to get various illnesses. Perlegen has just announced a large collaborative effort to search for genetic variations that are risk factors for type 2 diabetes.

Perlegen will use human genetic variations (single nucleotide polymorphisms or SNPs) it has discovered and its high-density oligonucleotide array-based SNP genotyping capability to assist researchers from around the world in intensifying their search for genes involved in a disease that affects approximately 15 million people in the United States and millions more around the world. The study includes geneticists at the University of Michigan, the University of Southern California, the National Public Health Institute of Finland, and the National Human Genome Research Institute in Bethesda, Maryland. During the past nine years, by studying the DNA of type 2 diabetes patients and their families, the FUSION group (Finland-United States Investigation Of Non-insulin-dependent diabetes mellitus genetics) has narrowed its search for the genes that play a causal role in the disease to certain areas of the human genome. An area of particularly high interest falls on the long arm of chromosome 6. Now, with help from Perlegen's scientists and the company's innovative genotyping technology, the FUSION group hopes to discover the gene or genes in that region.

What I find curious about their approach is that they used only 50 people as DNA sample donors. Surely it is not possible to find all the genetic variations that matter by looking at just 50 people. There are genetic sequence variations that show up in less than 2% of the population and hence one would expect one would need DNA from more than 50 people to identify all the SNPs that matter. Likely they took this approach for cost reasons. Most medically important SNPs are in the 1.7 million they identified and this is a cost effective way to search for most medically significant SNPs.

Update: Perlegen can identify all the unique sequences (or only out of the 1.7 million SNPs they've mapped?) in an individual's genoome in 10 days:

In theory, that goal is within reach, because researchers can now scan the entire genome to look for DNA variants of interest. Perlegen Sciences, a closely held company in Mountain View, Calif., recently announced that it can parse a person's genome in about ten days using so-called DNA chips--an astounding advance, given that it took an international army of scientists all of the 1990s to create the first draft of the human genome.

Update: Keep in mind that just as DNA sequencing technology advanced to make SNP detection faster and cheaper for Perlegen's recent work it will continue to advance and SNP detection costs will probably fall by orders of magnitude in the next ten years. Many more research groups and companies will be able to collect of SNP maps of larger groups of people for less money in the future. A company that seeks to make money off of SNP maps of smaller groups of people had better find useful SNP variations fairly quickly if they want to make a profit off their information. Collecting that information will only become cheaper going forward. If you want to get a sense of just how rapidly biotech assay tools are going to advance then read the FuturePundit Biotech Advance Rates archive.

Share |      Randall Parker, 2002 October 28 11:22 PM  Biotech Advance Rates

Post a comment
Name (not anon or anonymous):
Email Address:
Remember info?

Go Read More Posts On FuturePundit
Site Traffic Info
The contents of this site are copyright