Microchip Biotechnologies said last week that it has signed up its first early-access user for its first product, an instrument that automates DNA sample preparation for Sanger sequencing.
The instrument, called Apollo 100 Star, automates sequencing reaction setup, cycle-sequencing reactions, and sample cleanup, and can lower reagent and labor costs.
The Dublin, Calif.-based company said it plans to begin selling the system, primarily to sequencing core facilities, before the end of the year.
According to Barney Saunders, MBI’s chief commercial officer, the instrument has two main benefits: It uses small reaction volumes, which helps reduce reagent costs; and it eliminates manual steps, which lowers labor costs and reduces operator errors.
He said the company expects the system will reduce the cost of sequencing “for most centers.”
Apollo 100 Star integrates a liquid-handling robot, a thermal cycler, and reusable microfluidic chips. Users provide the system a 96-well microtiter plate containing the DNA samples, which the robot loads, together with sequencing reagents, onto four microfluidic chips, each of which processes 24 samples in parallel. These chips sit on a thermal cycler.
Samples and reagents are mixed inside the chips and move into a reaction chamber where cycle-sequencing reactions generate the labeled chain-terminated DNA fragments. The reaction products are automatically cleaned on-chip using paramagnetic beads, and the robot transfers the purified sequencing products back into a microtiter plate, ready to be loaded onto a capillary sequencer.
The instrument has been developed for use with ABI’s sequencing chemistry and Beckman’s Agencourt Solid Phase Reversible Immobilization beads for product purification.
Reaction volumes in the microchip are 1 microliter compared with the 10- to 20-microliter reactions used in a typical sequencing reaction, MBI said. The company is still optimizing the amounts of sequencing and cleanup reagents required, but Saunders said it will be at most half of what is needed for a typical sequencing reaction.
The company is also still determining the optimal amount of DNA template to be used but said on its website that the system “reduces the amount of precious sample needed in the process.”
Though the entire process, depending on cycling reaction protocols, typically takes 3 to 4 hours — which is only slightly less time than if the device were not used because a large part of it is the cycle-sequencing reactions — Saunders said Apollo 100 “free[s] up a lot of very valuable labor” that can be devoted to other tasks.
The instrument is designed to process two 96-well microtiter plates per day, or up to 50,000 samples per year, though a third plate can be run overnight, Saunders said.
According to a 2006 survey of 60 DNA sequencing facilities, conducted by the DNA Sequencing Research Group of the Association of Biomolecular Resource Facilities, the number of sequencing reactions run by a facility per year ranges from less than 10,000 to up to 500,000, with a median of 25,000 to 50,000 reactions.
How much labor the instrument saves, Saunders said, is “very variable” and depends on to what extent a lab has already automated its sample-prep process. Some labs already use robotics for the back-end cleanup process, for example, though the first part — setting up the sequencing reactions — is usually performed manually, and “nobody has built a microfluidic device that is married with a standard robot to do both processes,” according to Saunders. He estimated that overall, the system will halve the amount of labor required.
“The robustness of the platform will have to be demonstrated.”
For example, he said, the instrument could be interesting to core facility managers thinking about installing a new high-throughput sequencing platform, some of which require significant manual sample preparation. The Apollo 100 would allow these labs to shift labor from Sanger sequencing to the new sequencing platform. “They can get equipment grants, [but] it’s very hard to get funding for more labor,” Saunders maintained.
According to interviews with three sequencing core facility directors, the instrument could interest some labs. They were not sure, however, whether the instrument will reduce the overall cost of sequencing, which consists not only of reagents and labor, but also instrument depreciation and service, as well as overhead costs such as utilities.
“I’m sure there will be core facilities interested in this instrument because of the (potential) lowered cost and the ability to automate the process,” Peter Schweitzer, director of DNA sequencing and genotyping lab at Cornell University, told In Sequence by e-mail last week.
He said his facility might be interested in Apollo 100 if it can cut reagent costs “dramatically,” and if it enables researchers to use smaller amounts of DNA starting material. “The benefit of lowering the input DNA amounts might be something that’s really attractive.”
At the moment, his facility spends approximately $0.92 per reaction on sequencing and cleanup consumables, though “the bulk” of sequencing costs are associated with instrument depreciation and service contracts, other infrastructure costs, and labor, he said.
Also, if Apollo 100 could work with 384-well plates, “this would increase its attractiveness,” he said.
According to Schweitzer, the potential labor savings afforded by automation are less important for his facility “because we have an automated pipeline already” that uses conventional liquid-handling systems. Several of these systems, such as Beckman’s, already offer protocols for automating the reaction setup or the cleanup process, he noted.
Also, at least one other company, Parallabs, offers a platform, called Parallab 350, that automatically performs 96-cycle sequencing reactions in 0.5-microliter reaction volumes, and includes the cleanup step.
“The robustness of the [Apollo 100] platform will have to be demonstrated,” said Schweitzer, who said that it is “eerily reminiscent of the Parallabs instrument.”
Others believe that automation could be a benefit in some settings. “I can see the merits in a small operation where the sequencing technician is multitasking,” said Kevin Knudtson, director of the DNA facility at the University of Iowa. “This is certainly going to free their hands to get involved in other needs [besides setting up sequencing reactions] that might be provided by the core.”
According to Knudtson, “in an operation where people are dedicated to sequencing, and that’s what they are doing, I am not seeing that as being of great benefit,” especially if the facility’s sequencing throughput is large and technicians would use extra time to set up more plates, because the instrument, which cannot process more than three plates per day, would become the bottleneck.
He declined to comment on potential cost savings because the exact costs for running the Apollo 100 are not known yet.
MBI recently signed on its first early-access user for the station, the Canadian Center for DNA Barcoding at the Biodiversity Institute of Ontario at the University of Guelph (see In Sequence 4/8/2008
), which will receive the device early this fall.
The company said it would like to add around four more early-access sites before it fully launches Apollo 100 before the end of this year. MBI plans to sell the system for $60,000. The catalog price for a microchip, which can be used up to 50 times, will be $800, which includes reagents needed to re-coat the microfluidic chambers after cleaning.
MBI said it is also developing an automated sample-preparation device for pyrosequencing in collaboration with Mostafa Ronaghi at Stanford University (see In Sequence 3/4/2008). That tool is slated for commercialization within the next two to three years, the company said.