By Julia Karow
This article was originally published Oct. 26.
Hoping to capture a piece of the rapidly expanding market for targeted resequencing, Biodiscovery, a small microarray maker in Michigan, is expanding its business into capture arrays and probe libraries for targeted next-generation sequencing.
Based on low-cost microarray fabrication technology developed in the lab of Erdogan Gulari at the University of Michigan, the Ann Arbor-based company was founded in late 2006 with $2.2 million in funding from the state of Michigan and a $300,000 grant from the National Human Genome Research Institute. Biodiscovery, which currently has six full-time employees, has been operating since mid-2007 under the name Mycroarray.com in order to avoid confusion with El Segundo, Calif.-based microarray software developer BioDiscovery.
Although Mycroarray.com faces considerable competition from Agilent Technologies, Roche/NimbleGen, Febit, and other providers of capture arrays and probe libraries for targeted sequencing, it is confident that its flexible design and low cost will make it an attractive alternative.
Until now, most of the firm's business has come from custom microarrays for gene expression studies, mostly for bacteria. It also offers hundreds of catalog arrays for bacteria, almost 50 for archaea, and about a dozen for eukaryotes. "We have found a really nice niche market in the bacterial world," Jean-Marie Rouillard, the company's chief scientific officer and one of its co-founders, told In Sequence, banking on its design flexibility and cost-competitiveness compared to other players. Adding capture arrays and probe libraries for targeted resequencing is "a natural expansion of our business," he said.
Mycroarray.com synthesizes oligonucleotides directly on the chip, using standard phosphoramidite chemistry. Each synthesis cycle involves the removal of a protection group by an acid, a step that the firm has modified by using a photo-generated acid, which it produces by shining light on the appropriate spots on the array. Instead of physical masks, it uses a digital micromirror device to project light on the array spots, similar to digital photolithography technology used by NimbleGen, Rouillard explained. However, its chemistry allows it to produce longer probes than NimbleGen's — up to 150 nucleotides.
The company currently offers catalog gene expression arrays with up to 90,000 spots, although it has already reached up to 500,000 spots per slide internally.
Prices for its arrays range between $40 for 5,000-spot arrays to $260 for 90,000-spot arrays. "We are probably cheaper by at least a factor of two on many products" from competitors, Rouillard estimated. The company also does not require order minimums and does not apply customization fees. "This is why we especially focus our market on customized products," he said.
At the moment, customers can order custom arrays for gene expression analysis as well as for targeted sequence capture, and Mycroarray.com offers a design service for both. In a demonstration project, the firm has shown that it can use its arrays to selectively capture DNA from a single yeast chromosome, measured by how much of the captured material bound to a detection microarray. Because it currently does not have access to a next-generation sequencer, it has not yet validated the capture efficiency by sequencing but is seeking a collaboration with a core facility to be able to do so.
However, Rouillard said he believes array-based sequence capture will likely "disappear in a year or two" because "it's so much easier, so much more efficient to do it in liquid phase."
Thus, the company is also working on an in-solution capture product, similar to Agilent's SureSelect, which it wants to launch by the end of the year. The plan is to offer custom libraries of biotinylated RNA baits, but with 90,000 probes instead of Agilent's 55,000 different baits. To release the oligos from the chip, Mycroarray.com uses a proprietary technology to select only oligos from the center of the spots, thus avoiding "junk sequences" that may develop at the edge when using the phosphoramidite chemistry, according to Rouillard.
The firm is currently working on proof of concept for its in-solution product, he said, and has not yet decided whether to offer probe libraries with protocols or a full kit that also includes reagents. The first libraries will probably contain up to 90,000 120-mer probes.
Regarding array-based capture, "one could argue that you have a higher density of probes on NimbleGen arrays, for example, so you could capture more sequence," Rouillard said. "But as the price of making a capture library in solution is going to decrease, especially when we are going to enter the market — we may cut the price by a factor of two — it will be more cost-effective to make a couple of libraries and mix them together to capture in solution rather than using a large array," he said.
So far, the company does not have partnerships with vendors of sequencing technology and currently plans to offer its capture technology for use with any sequencing platform.
Rouillard said he is unfazed by existing competitors, including Agilent, which he said will be its "closest competitor" because it also offers long probes. "It was the same problem for us when we started the microarray business," he said. "We have a lot of competitors, but we are growing our market share fast because we are very cost-effective because of the chemistry we are using … and since we are very flexible, we can offer something that was not available on the market before.
"We really believe that we can bring the same flexibility and the same low cost to targeted capture for sequencing."