By Julia Karow
This article, originally published Sept. 21, has been updated from a previous version to include additional information from a company official.
Pacific Biosciences said this week that it plans to focus on cancer and infectious diseases as initial application areas for its single-molecule real-time sequencing platform, noting that these areas are best suited to take advantage of the system's long reads and short cycle time.
In preparation for the full commercial launch of its instrument during the second half of 2010, the company has begun to build a sales force, and to scale up in the areas of manufacturing and operations.
PacBio is also working on additional applications for its SMRT analysis platform, to be introduced over time, including methylation sequencing, direct RNA sequencing, and protein translation.
The firm further disclosed that it is collaborating with six early-access customers, including Monsanto and the Scripps Institute, on a variety of sequencing projects. It will soon finalize the list of institutions that will receive an early commercial version of its SMRT sequencer during the first half of next year, which will likely include Monsanto and large academic genome centers.
At the UBS Global Life Sciences conference in New York this week, PacBio CEO Hugh Martin said that the company plans to focus initially on sequencing applications in cancer and infectious disease research, where it sees the strongest differentiation from its competitors.
The platform's long reads — expected to be at least 1,500 bases at launch next year — are especially sought after by cancer researchers, he said, and its short run time, or cycle time, of 15 minutes makes it attractive for infectious disease applications that require a fast answer.
At the moment, Martin said, the company has six collaborations underway that cover a variety of sequencing projects, including candidate variant discovery, agricultural biology, medical sequencing for cancer, and infectious diseases. Collaborators include three genome centers, two academic centers, and one company, which he later identified as Monsanto, an investor in PacBio's last funding round.
Also among the collaborators is Nicholas Schork's team at the Scripps Institute in San Diego, which uses PacBio data to study drug-resistant bacteria, according to PacBio Vice President of Marketing Martha Trela.
During the first half of next year, the company plans to ship on the order of 10 instruments — an early version of the commercial platform — to a number of institutions. Trela told In Sequence last week that the firm plans to "finalize the list" of these customers within the next couple of months. Monsanto is "very interested in being a customer," she said, and the firm "will be in discussions with big genome centers about their level of interest and criteria for those test sites."
In terms of customer types, Martin said the company will first concentrate on large genome centers and academic institutions, followed by pharmaceutical companies and "smaller individual customers," and finally on clinical diagnostics customers.
The company has already helped researchers with 45 grant applications that mention PacBio's platform "as the preferred piece of hardware," he said, and the firm believes it will be able to benefit from stimulus funding.
However, Martin cautioned that there are "only so many instruments" PacBio can manufacture next year and in 2011, and said that it will need to "think about customer diversity" so not all instruments go to one type of customer, such as genome centers.
[ pagebreak ]
He said the firm sees "quite a big opportunity" in the biotech and pharma market as well as the applied and industrial market, where traditional Sanger sequencing still dominates today.
According to Martin's forecast, the DNA sequencing market will grow from approximately $1.4 billion today, more or less evenly split between Sanger and next-gen sequencing, to about $1.8 billion by 2013, with the growth coming from next-gen sequencing.
Facing the Competition
Martin also squarely addressed the competition, and where the company believes it has an edge.
Roche/454's Genome Sequencer FLX will be "the product most directly impacted by us," he predicted. Users of the 454 platform have been mainly interested in its long reads, compared to other next-generation sequencing platforms, he said, and PacBio will be able to provide even longer reads at a lower cost.
Compared to Illumina's and Life Technologies' short-read platforms, which he said have “roughly the same kind of performance profile,” even though Illumina’s Genome Analyzer appears to be dominating in the market, PacBio will offer a new type of sequence data that will be fundamentally different from either of the two. After PacBio's platform launches, he predicted, there will likely only be room and sufficient "academic drive" to support one other next-gen sequencing platform.
Martin also addressed common criticism of its platform by others. Existing next-gen sequencing vendors say, for instance, that PacBio's accuracy is inferior to that of their respective platforms, but he said that is only true for raw read accuracy, not for accuracy of "finished" sequence. He said that over the next two months or so, the company will start an "education process," through a series of papers, to inform researchers about the nature of its data.
He also addressed claims that PacBio's throughput will never be as high as that of second-generation sequencing systems because of the optical requirements for real-time sequencing. While that is true in general, he said, people will be "very surprised" when the company reveals its initial throughput next year, which relies on recent "breakthroughs" in sensor technology that he said the company has heavily invested in.
The throughput of the first instrument version can be increased several-fold in two ways without hardware changes, he explained. First, the company could switch the current DNA polymerase, which runs at 3 to 5 bases per second, for one that works at 15 bases per second, he said, noting that PacBio is already running sequencing experiments internally at 10 bases per second. In addition, the company is working on ways to actively fill the zero-mode waveguides with polymerase, thus increasing their occupancy from currently one-third to 80 percent to 90 percent.
Long term, Martin said, the company wants to develop an upgraded instrument that can generate 100 gigabases of sequence per hour. Reads could grow to between 50,000 and 100,000 base pairs, and the company has already developed a polymerase that can synthesize DNA at 25 bases per second, which he said could be increased to 50 bases per second. The company is already working on a chip with a million zero-mode waveguides, which it plans to introduce in 2013 or 2014. Martin also mentioned that the company might not always use fluorescent labels but may instead switch to other types of labels in the future.
Although PacBio has not revealed all specs for its first commercial instrument, Martin said it will have reads of at least 1,500 bases and a "time to result," or run time, of 15 minutes. A run will cost approximately $60 in reagents, and an experiment — including sample preparation, which does not require amplification — can be completed in a single day, which no other next-generation sequencer except for 454’s achieves today.
PacBio has said previously that the first commercial version of the instrument will cost as much as current high-throughput sequencers, or on the order of half a million dollars, though Martin noted this week that the company will not price the instrument "right on top" of Illumina's Genome Analyzer.
The company expects that users will purchase on the order of hundreds of thousands of dollars worth of consumables per year for the instrument, according to Trela, similar to what Illumina has said its GA customers spend. A single sequencing experiment will cost "no more than in the low hundreds," in total, according to Trela.
[ pagebreak ]
Asked what the biggest near-term hurdle for the company is, Martin said that the instrument is "a very complex piece of equipment" with more than 3,000 components, and the company is currently most focused on executing the assembly and quality-control of its instruments.
Looking Beyond DNA Sequencing
Long-term, the company is also banking on other applications for its platform besides straight DNA sequencing. Trela said that the firm knew from the start that its technology, the principles of which were first published by Cornell University researchers in 2003, was amenable to study different types of single-molecule interactions, but it decided to focus on DNA sequencing first.
Since then, "we have also had a lot of people come to us and say, 'I'd love to look at RNA directly, or at ribosomes, or other types of applications,'" she said. DNA methylation analysis, in particular, "has been one of the applications the scientific community has shown a tremendous amount of interest in, and we think it could add much value," she said.
Partly in response to these requests, the company decided to work on additional applications for its platform, both internally and through scientific collaborations. New applications include methylation sequencing, direct RNA sequencing, and the analysis of protein translation. For the latter — observing ribosomes as they translate messenger RNA into protein — the firm is collaborating with an undisclosed research group at Stanford University.
It is unclear yet whether any of the new applications will be available when the platform is first launched later next year for DNA sequencing. "We'll develop them as quickly as we can," Trela said.
At last week's Personal Genomes conference at Cold Spring Harbor Laboratory, PacBio Chief Technology Officer Steve Turner presented early results for DNA methylation and RNA sequencing (see related article in this issue). He said that direct RNA sequencing will be available within a year of the platform's launch.
In preparation for the full launch of the instrument in the second half of 2010, PacBio has also started to build out its commercial arm, scaled up manufacturing and operations, and begun to assemble a sales force. The firm just hired Pat Brooks, a former vice president of sales for Affymetrix, as vice president for North American sales. It also added Randy Livingston, chief financial officer of Stanford University, to its board of directors.
Deciding in June to build out its own commercial operations, Martin said, was "effectively a vote by the board to become an independent public company." He said PacBio has "a strong desire" to go public and will "think about an IPO" next year if the market conditions are right.
The company has approximately 290 employees, and "we are continuing to grow and add," Trela said, but did not put a specific number on how much the staff is expected to increase in the near term.
PacBio expects to be profitable by mid-2012, and to spend $350 million in total before it becomes cash-flow positive. To date, it has raised $266 million in venture capital.