Danaher Motion’s Dover business has been taking orders for the “Polonator,” the next-generation sequencer it has been co-developing with George Church’s lab at Harvard Medical School, and hopes to ship its first production units by the end of the year, In Sequence has learned.
The instrument, which can support different biochemistries, is competing with Helicos BioSciences’ HeliScope to become the fourth next-gen sequencer to reach the market. But Danaher’s distribution model differs markedly from that of 454 Life Sciences, Illumina, and Applied Biosystems: the company will not sell flow cells and reagent kits, or provide support for issues unrelated to the instrument’s hardware. The Church lab, which is still assessing the instrument’s performance, will be providing sequencing protocols and open-source software to run the machine.
“This is different than the way things have been done before,” Kevin McCarthy, chief technology officer of the Dover business, told In Sequence last week. “It’s meant to be a modular, open source platform” that could even be used for applications other than sequencing.
In fact, in a written description of the instrument that was circulated at a recent core director’s meeting (see In Sequence 11/20/2007), Danaher bills the instrument as “the G.007 automated fluorescence microscope,” rather than a sequencer, that could also be used, for example, for live cell fluorescence imaging. “We will market it for anything; we wish to sell as many as possible,” McCarthy said. “It’s multi-purpose, although certainly its first, and probably the most important, application will be genome sequencing.”
Though the system can accommodate different sequencing biochemistries, it will initially support protocols from the Church lab that employ polony sequencing-by-ligation, lending the instrument its “polonator” nickname.
G.007 refers to the design level and year of the system, and the first 2008 version will be called A.008. Each version is “design-locked” but can accommodate upgrades, and each version can be upgraded to the next. McCarthy said he expects about two or three upgrades per year.
The system’s performance as a sequencer is still a big unknown, and is currently being evaluated by the Church laboratory using its sequencing-by-ligation biochemistry on a prototype F.007 version that the lab recently acquired (see In Sequence 11/6/2007).
According to a talk Church gave in October (see In Sequence 10/16/2007), the instrument has an output of 10 gigabases of raw data per run, based on a read length of 28 bases, and the cost of sequencing is about $3,000 per gigabase. Next year, the output efficiency is expected to improve twofold and the cost six-fold. Greg Porreca, one of the developers in the Church lab, told In Sequence last week that “those figures are still what we anticipate,” but he did not provide a timeline for when actual performance data will be available.
McCarthy, who characterized the Church team as “a brilliant group of engineers and scientists” believes that the instrument “is going to be a higher throughput system, probably higher accuracy, but more particularly, dramatically lower cost,” than those from competitors.
According to the Danaher description, the company expects “an aggressive program of improvement in performance; several major changes are already planned, but we expect others, including ones we could not anticipate, to arise from feedback from a growing user community.”
McCarthy declined to provide details about the planned changes but said that “the upgrades don’t imply a weakness on the part of the G.007, but we think we can take it up even further.”
The instrument will cost just under $150,000, slightly more than earlier estimates of around $130,000 because “some substantial additions have occurred in the last month which make the instrument much more powerful,” according to McCarthy, including a more powerful computer. Pricing for flow cells and reagent kits is not available yet, according to Porreca.
Reagent kits will contain off-the-shelf rather than proprietary reagents and will be provided by an unnamed third-party vendor. The Church lab has said in the past that it is working with Enzymatics, a company based in Beverly, Mass., to cut the cost of the enzymes required. According to McCarthy, the researchers have also “set up relationships” with oligonucleotide providers.
Flow cells will come from yet another unnamed vendor, though other companies might join. “All the designs are open, so I foresee multiple vendors as the project grows, encouraging healthy competition,” Rich Terry, another project leader in the Church lab, told In Sequence last week.
“This is different than the way things have been done before.”
The Church lab will provide tested sequencing protocols, as well as operating software, which is open source and allows users to submit changes. “As we have seen with [computer] operating systems, that permits some pretty fast [improvements]. You get a lot of minds thinking about it,” McCarthy said.
When it comes to service and support, Danaher will only take care of problems relating to the instrument hardware. “If you are noticing, for example, that on your 20th base read the signal is weaker than you wanted it to be, that’s a protocol problem, and we are not going to be able to help you with that,” McCarthy said. “We are responsible for collecting the photons and providing the images.”
With regards to protocol-related issues, “since the protocols are open, users (with our guidance) should be able to troubleshoot any problems that arise,” Porreca said.
Danaher will also provide instrument training as an additional service, but it is unclear right now whether that will include an actual sequencing run.
Despite the uncertainties surrounding the instrument, Danaher has already seen considerable interest in the platform. “We have probably 15 people who want [the instrument],” McCarthy said. “Many of those are influential first adoptors who would be in a position to provide a good review of the instrument and inform the community as to an independent assessment of its value.”
He hopes the company will ship between two and four instruments by the end of the year, or by mid-January at the latest, to the Church lab and to other undisclosed sites. Right now, the plan is to produce five instruments per week, though “that’s easily scaleable in either direction,” he said. “If we don’t see the demand we think might be there, we can easily take it down. If we need to take it above that level, we can also do that.”
The sequencer is not the first foray of Danaher Motion’s Dover business into life sciences. The unit, which has annual sales between $45 million and $65 million, specializes in high-precision stages for a variety of markets, including the semiconductor industry, data storage, flat panel displays, general factory automation, and life sciences. “We are very good at very high throughput, very precise motion,” McCarthy said.
Life science applications account for about a quarter of the Dover business. For example, the company provides components for an imaging microscope for the automated analysis of Pap smears, sold by Cytyc, a subsidiary of Hologic. “The new G.007 automated fluorescence microscope is in essence a larger version of our [component in the] Cytyc platform, while adding several exciting new features,” Danaher’s instrument description reads. Interestingly, Cytyc was founded in 1987 by Stan Lapidus, CEO of single-molecule sequencing company Helicos BioSciences.
McCarthy, who has an undergraduate degree in physics from the Massachusetts Institute of Technology, is a veteran of the motion stage business. For 25 years, he worked for New England Affiliated Technologies, a company he co-owned. He and his partner sold that business in 1999 to a company that Danaher Motion acquired a year later. In 2002, Danaher added another company, Dover Instruments, to the unit.
But the expertise of the Dover business goes beyond stages. “As we interact with opportunities, we often grow the boundaries,” McCarthy said. “We are not just supplying the motion subsystem, but we are supplying closer to complete systems.”
In the case of the G.007, the company will provide the entire system. McCarthy first met with the Church lab in April to talk about providing better stages. At that time, he said, the lab was considering an entirely “open-source” design where users would play some role in assembling the system.
McCarthy said that the two parties soon recognized that the “average biologist” might not be interested in this model, and agreed over the summer that Danaher would manufacture the entire machine, and play a role in the design and development of the instrument.
“We are excited about helping them realize their vision" of “bringing second generation sequencing into the mainstream, in a way that is both open and affordable,” McCarthy said.