Ernest Kawasaki, head of the microarray facility of the Center for Cancer Research at the National Cancer Institute at Gaithersburg, Md., is the kind of customer that the microarray industry covets: His facility is among the heavy-duty users of self-spotting microarray technology.
Last week, when Applied Biosystems announced that it was planning to release by the end of the year a single microarray containing the whole human genome and instrumentation to read it (see below), he was listening.
“I would use it,” Kawaski told BioArray News. The spotting schema used at the NCI facility can print arrays of 20,000 to 40,000 spots currently, but the technology falters at the upper end of the range, he said. A rigorously manufactured product would solve that problem, and Kawasaki said he knows it is just a matter of time before his days of running a facility that prints its own microarrays are over.
“In a few years from now, nobody is going to have their core facility making arrays,” he said. “I’ll be out of a job in three or four years.”
More likely, Kawasaki’s job will transform — just as the industry does.
Kawasaki, and scientists like him, represent the revenue sweet spot of the microarray industry that is, by some estimates, at least 50 percent home brew: Industry watchers predict this group will eventually be won over by the industrial- scale production of standard products that are extremely sensitive and accurate, and generate reproducible results.
The FDA has said, in many ways recently, that it is not ready to give its regulatory blessings to data derived from this tool.
But, events of the last three weeks — regulatory letters from the FDA, the entry of a new player, and the launch of a race to create a new product — indicate a new dynamic marketplace for this eight-year-old technology whose commercializers are looking for a bridge between research laboratory bench tops and the lucrative clinical market where, some day, tests conducted on microarray-based technology will save money, and lives. Higher- density arrays are just a milestone and a measure of innovation and progress in the commercialization of this technology.
ABI’s press-release-fueled entry into the market last week was like a swimmer bellyflopping into a pool, a less-than-graceful entry marked by an announcement of plans for a new product but with no corroborating detail to support claims of innovation and progress.
The entry of Agilent and NimbleGen this week with announcements (see page 10) of whole-genome microarray programs makes it a dash to produce, and sell, a whole- human-genome, single-microarray product, and brings a notion that paradigm shift is in the air despite the fact that Affymetrix, the industry leader, and Amersham, last year’s newcomer, have opted not to dive in — just yet.
There are others too. Earlier in the month, Taiwan’s Industrial Technology Research Institute, the private-public collaboration that can take much credit for producing the country’s powerful semiconductor manufacturing industry, told BioArray News that its spinoff, Phalanx Group, was preparing to manufacture its own high-density microarrays, targeting a whole- human-genome single microarray by the end of the year.
Meantime, the exact content of the human genome, while officially sequenced, is dynamic, changing weekly as discovery continues. And, there are also proteins to characterize, not to mention the sugars, lipids, small molecules, and other components of the metabolome that scientists are trying to understand in a systematic fashion.
“The whole genome array has a very fuzzy definition,” Chungcheng Liu, director of the molecular biomedical technology division of ITRI, told BioArray News. “We don’t know exactly how many genes are in the human, definitively, and we don’t know how many transcripts can be turned into protein. Without that, nobody can claim ‘I have a whole genome array,’ including ourselves. It is a metaphor.”
Big pharma customers — the likely first customers of whole-genome microarrays — are also demanding more than just gene probes.
They really want chips with all of the RNA that is transcribed, or SNPs, or the proteins, a highly-placed pharmaceutical industry genomics executive told BioArray News.
Pharmaceutical researchers, the executive said, would rather look at all of the proteins rather than the messages. “Concordance is an issue with RNA.”
Having all the post-translational modifications would be useful too, the executive said. “But proteomics is not that advanced — it’s really not an ‘omic yet, while transcription is truly an ‘omic. With 2D gels, you are only looking at 4 percent of the genome. You are just putting your toe in the water.”