Salt Lake City Bioscience plans to commercialize a new technology platform developed by the University of Utah.
Called real-time microarrays, which allow users to monitor chemical reactions on the surface of a chip in real time, the products are said to provide "a much more accurate description" of what is in a particular sample.
Portland Bioscience President Steve Benight told BioArray News last week that the company's subsidiary, Salt Lake City Bioscience, is interested in bringing the new technology to market.
"We have a considerable amount of research to do, just in calibrating the system," Benight said. "We can multiplex, but not at a high level yet. But we are [having] preliminary talks with a few smaller diagnostic companies and they have expressed interest."
Benight said the technology could become commercially available within a year "at the earliest."
The real-time microarray platform was developed by Steve Blair, director of the center for microarray technology in the department of electrical and computer engineering at the University of Utah in Salt Lake City.
It is designed so that researchers can monitor the chemical reactions on its surface in real time, Blair told BioArray News.
"Rather than doing an endpoint measurement, which is most common in microarray experiments, with real-time microarrays, we monitor reactions as they occur so we can generate data continuously over a certain period of time," he said.
That leads to more accurate data, Blair argued.
"Rather than having a single data point, we have a whole time course," he said. "Fitting that time course to a mathematical model, so it's a model-based approach to interpreting the data, we can get a much more accurate description of what was in our sample, and what reacted with a particular spot."
The technology has been at the center of a five-year National Institutes of Health grant, entitled "Real-time heteroplasmy analysis on microarrays." Begun in 2008 and scheduled to expire next year, the project has received $1.7 million in federal support to date.
According to the grant abstract, the developers believe that the real-time microarray platform overcomes the inability of conventional SNP microarrays to offer quantitative results.
"Current SNP analysis is not quantitative because of imperfect molecular recognition, such as cross-hybridization, and pseudoequilibrium analyses performed with microarrays, limiting their use," the abstract states. "The microarray techniques attempt to compensate via excessive redundancy, leading to massive quantities of inaccurate and irreproducible data."
One "key task" that the developers said they believe the platform can address is heteroplasmy, or the presence of a mixture of more than one type of mitochondrial DNA within a cell or individual, which is a factor in the severity of mitochondrial diseases.
Their stated goal is to "develop, prototype, and commercialize real-time and quantitative heteroplasmy research tools" based on technology licensed from the U of Utah. The proposed method of performing real-time SNP microarray analysis calls for "monitoring non-linear binding kinetics of known competitors in the presence of unlabeled targets."
Benight said the technology "provides an attractive alternative" to conventional SNP arrays, and argued that the resolution has a "higher level discrimination for hybridization events including those including SNPs and variants of disease interest."
Still, he said, the platform has a ways to go before it can be used in a commercial test.
"From the research side, we have to demonstrate our multiplex and probe sets with ideal targets before we can go to real sample," said Benight, adding that the firm's assay will eventually be tested by an as-yet undisclosed third party that specializes in mitochondrial genomic analysis.
[ pagebreak ]
Another issue is the number of probes to optimize for use on the array. He said that platform developers need to decide "which probes will be most efficient in the format that we need for different disease loci that we are trying to identify." In coming months, the developers plan to calibrate the real-time array "with a couple of ideal targets," he said.
Getting diagnostics firms interested in the technology is another challenge.
"In my experience, the diagnostics companies that would be interested in this in a major way, to utilize the technology and get it commercialized, have already taken a swim in the microarray pool and had marginal success to the point where their investors have been burned," said Benight. "That being said, we are making progress on the assay, and we do have a commercial partner who is willing to test it."
Blair said adapting the platform for use in diagnostics is a "nice match" for such medium-density arrays, which average about 500 probes.
"It's not just discovery," he said. "We have known loci that we are trying to very accurately detect."
In terms of the heteroplasmy project, Blair said that the "real challenge" is detecting allele-frequency variation.
"You may have mutations or SNPs at a fairly low frequency relative to wild type, but there are a number of studies out there that have shown that even these low-frequency variations are predictive of disease," he said.
Beyond heteroplasmy, he is also developing real-time microarrays for methylation profiling.
"We take methylated DNA and hybridize it to an array that contains probes that are themselves methylated," he said. Proteins are then bound to the array in a second step. "Within a given sequence, we can tell you if it's methylated in one of any number positions. So it is very precise in terms of location."
While Blair's lab has mainly been printing the real-time arrays internally, he said the assay can work with catalog arrays, though cautioned that such chips have their limits.
"The one caveat with the method that we use to achieve real-time detection is that we have to get around the challenge of background detection," said Blair. "The entire solution covering the array is fluorescent, so we use a method of total internal reflection fluorescence to interrogate surface selectively, not excite the background."
The challenge with TRF is the fluorescence of the substrate, according to Blair. "A challenge for us in using off-the-shelf arrays is being able to achieve high enough signal-to-background detection," he explained.
For now Blair's team relies on quartz microscope slides and has the chips printed by an undisclosed external manufacturer. "We do the surface modification and send them out to be printed," he said, adding that he is considering adopting arrays fabricated by vendors like Agilent Technologies or Roche NimbleGen, but has not yet made any definitive decision.
Have topics you'd like to see covered in BioArray News? Contact the editor at jpetrone [at] genomeweb [.] com.