A Wall Street Journal story on Dec. 21 vividly chronicles the battle between two rival technologies to chart genetic differences in clinical trials. Industry leader Affymetrix is losing ground and market share to Illumina. The story notes that the Illumina role in a key scientific paper on inflammatory bowel disease has sent ripples through other laboratories.

Here’s a quote:

“The two companies are battling it out to sell equipment for ‘high through-put genotyping’—a market now roughly $200 million to $350 million a year in sales and growing by 25% to 40% a year. Most of that is in gene chips and scanners used to perform DNA sweeps of thousands of people to find the genes linked to cancer, asthma, autism and other conditions.”

And another:

“Just over a year ago, Affymetrix rushed to unveil its new-generation chip that squeezed half a million SNPs on a single, two-chip product for genotyping. In its haste, it stumbled. The product was under-tested, and had flawed software. It took five months, a new software download, and delivery of new chips to fix the problem, according to Affymetrix.”

Even outside the laboratory equipment, there is interesting math that still needs to be developed and analyzed.

The software company Insightful is on the case. As of November, 2006, they have a $750,000 NIH grant to use least angle regression with complex genetic (and other) datasets. We’ll gladly confess we have no idea what least angle regression software (LARS) is, but it is apparently used when there are multiple confounding variables.

Rather than dig ourselves in deeper, here’s a quote from the Insightful news release:

“This project calls for Insightful to develop an easy-to-use and flexible LARS solution that may be used for medical diagnosis, detection of cancer, feature selection in microarrays, and modeling patient characteristics like blood pressure…. ‘Researchers and academics believe that LARS has great potential, but up till now the academic software available has been limited in scope and refinement,’ said Dr. Jill R. Goldschneider, director of research at Insightful. ‘We are excited by this opportunity granted by the NIH to develop robust and easy-to-use LARS software that will enable medical researchers to obtain high prediction accuracy and to obtain stable and interpretable results in high-dimensional situations.’ ”