Lab-on-a-chip developer RainDance Technologies is using a recent grant from the National Cancer Institute to develop an automated sample-preparation system for exon sequencing — one of several “killer applications” the company is developing for its upcoming Personal Laboratory System, CEO Jonathan Rothberg told In Sequence last week.
The $140,625 grant, awarded last month, will help RainDance develop a microfluidic chip that processes genomic DNA to enable thousands of exons to be sequenced simultaneously from a single sample.
The chip will be based on the company’s proprietary fluid-handling technology, which encapsulates biomolecules into individual micron-sized droplets that move through microfluidic channels that are preconfigured for a number of laboratory tasks.
The exon sample-prep chip is one of several laboratory applications RainDance is developing for the PLS that will come in the form of cartridges that Rothberg likened to floppy disks for a personal computer.
“You’d have a kit for whatever application you want, and the droplets are loaded exactly for your application,” said Rothberg, who is also founder and chairman of the company.
According to the grant abstract, the system will enable researchers to sequence thousands of different exons simultaneously from a genomic DNA sample “with 30 to 50 times coverage of each exon.”
The company plans to develop a microfluidic chip that will “combine genomic DNA on a droplet-by-droplet basis with a member of a library of thousands of primer pairs at rates of thousands per second,” according to the grant abstract. “The primer pairs ensure that a single exon is amplified in a given droplet during a polymerase chain reaction step.”
The product of that step is captured on a bead for sequencing, and then beads with less than 10 million copies of DNA are removed before the remaining beads are sent for sequencing.
“Successful completion of this project will allow one pooled-primer-pair-library emulsion to be used to economically sequence thousands of individual exons from an almost unlimited number of genomic, (including tumor), DNA samples,” the abstract states.
RainDance plans to release the first version of the PLS to early-adopter labs by the fall, and hopes to have several application-specific cartridges ready to ship with the system. Rothberg declined to provide details on which applications would be ready by that time. The company hopes to formally launch the PLS system in the first quarter of 2008, he said.
In addition to sequencing, Rothberg said the company has targeted kinase screening, RNAi screening, and functional flow sorting for some early applications, but has not disclosed a launch timeline for any of them.
While the RainDance platform is envisioned as a general-purpose laboratory instrument, it was perhaps inevitable that one of its first applications would be in sequencing. Rothberg, after all, founded CuraGen and 454 Life Sciences before founding RainDance in 2004.
While Rothberg was careful to stress that the exon-sequencing cartridge is “just one of our killer applications,” he noted that the company views DNA sample prep as a prime market opportunity, particularly within the context of next-generation sequencing technologies.
The NCI grant abstract describes the exon-sequencing chip as a front end for the 454 platform, but Rothberg didn’t rule out the possibility that it could work with other sequencing platforms.
Acknowledging his bias toward 454, he said that the RainDance sample-prep system is “ideal for the 454 platform.” Nevertheless, he noted, “nothing is stopping us from prepping for … capillary ABI, the new SOLiD, and doing prep for Solexa. Nothing is out of the question.”
Shortcut to the $10K Genome
The system promises to greatly simplify sample prep for the 454 instrument, which currently takes “a skilled operator” to perform, Rothberg said.
He said that RainDance decided to focus on exon sequencing rather than whole-genome sequencing because it serves the real-world DNA-analysis needs of most researchers today — without having to wait for the $1,000 genome.
As an example, Rothberg cited 454’s recent project with the Baylor College of Medicine to sequence Jim Watson’s genome for under $1 million (see In Sequence 06/05/07).
Nothing is stopping us from prepping for … capillary ABI, the new SOLiD, and doing prep for Solexa. Nothing is out of the question.”
“When we actually looked at the data, we only looked at 1 percent of the genome, and that 1 percent is all the coding regions,” he said. “It’s great to be able to sequence the first individual for a million bucks, but you realize that when you’re all finished with it, you only look at 1 percent of the data.”
Rothberg said that by focusing on exons, even with a million-dollar price tag for the whole genome, “if you just sequence the 1 percent, and you do the math, you’re at the $10,000 genome today.”
RainDance envisions targeted kits for researchers interested in all the exons in the genome, or those genes that are implicated in cancer, such as all the kinases.
“It's a neat idea,” said Jay Shendure, who has been working on a method to amplify tens of thousands of exons in a single reaction.
In an e-mail to In Sequence, Shendure, who will take up a post as an assistant professor in the department of genome sciences at the University of Washington this fall, said that while he is not familiar with details of RainDance’s system, the concept behind the technology seems sound.
“In principle, if individual exon amplifications are confined to droplets, this may result in normalized amplification of various exons relative to one another,” he said. “The catch is that it still requires one-by-one synthesis of primer oligonucleotides and therefore potentially a large upfront capital cost, but provided the amounts needed per reaction are small, the amortized cost could be very low.”
Richard Gibbs, co-director of Baylor’s Human Genome Sequencing Center, noted in an e-mail that the RainDance sample-prep system “looks like a good technology for allowing parallel handling of large numbers of samples at the ‘front-end’ of the sequencing problem.”
He added that Baylor is not currently collaborating with RainDance, but said that his group has had some discussions with the company and is “optimistic that we will be working with them soon.”
While Baylor and other large sequencing centers are the most likely candidates as early adopters for the RainDance system, Rothberg stressed that one of the company’s goals is to “democratize” sequencing “so it allows you to move from specialized places like genome centers to academic labs, commercial labs, and then diagnostic labs.”
Eventually, he said, “all of us are going to be sequenced, and it will be nice if it’s prepped on a RainDance machine because that fits right into an academic lab or a diagnostic lab, as opposed to having to worry about robots or more sophisticated things — which obviously [the Broad Institute’s] Eric Lander can do and Richard Gibbs can do — but if you want to make something much more general, much more accessible, you really want to put in a cartridge … and then you push a button on the screen and out comes your DNA prep.”
— Julia Karow contributed to this article