Illumina is seeing greater market demand for its sequencers than it had expected, and provided some details last week during its analyst day conference in New York.
The company had shipped 100 Genome Analyzers as of late August and wants to account for half of all installed next-gen sequencers by year’s end, according to John West, senior vice president and general manager of sequencing.
Illumina is also working on a number of technical improvements to increase the output and lower the cost of sequencing on its new platform. In addition, the company believes that starting in 2008, whole-genome sequencing applications will cause the sequencing market to grow beyond what it believes totals $1 billion today.
West reiterated that less than half of the company’s sequencing business today comes from large genome centers (see In Sequence 7/31/2007).
This summer, Illumina said that the Broad Institute had installed 16 instruments “of a possible 20” and that another single, unnamed customer had ordered 20 instruments. Last week, West said that this order also came from “one of the large sequencing centers.”
In addition to genome centers, the company has sold instruments to core labs, industrial laboratories, agricultural firms, biotech, and pharmaceutical companies, West said.
By the end of the year, Illumina hopes to capture at least half of the installed base of all next-generation sequencers. “We would expect a total installed base of all the companies that are participating in this market of over 200 instruments by the end of this year,” West said at the conference. “Illumina has set a target of having over 50 percent of that cumulative installed base by the end of this year.”
But according to West, the company has already hit that target. “As of a couple of weeks ago,” he said, Illumina already shipped 100 systems to customers. By comparison, 454 Life Sciences said it placed 20 units of its Genome Sequencer 20 in 2005, its first year of sales, and more than 40 sequencers in 2006.
But 454’s current numbers raise questions about Illumina’s estimate on the total installed base of next-gen instruments by year’s end. At a conference in May, 454 founder Jonathan Rothberg said that 454, a subsidiary of Roche, had a worldwide installed base of approximately 80 to 100 instruments at that time. He also predicted that the company would sell more than 100 sequencers this year, which would bring the total installed base of 454’s system alone to nearly 200.
A 454 spokesman told In Sequence this week that Rothberg’s estimate was “too low.”
In addition, ABI has shipped its SOLiD platform to several early-access customers but has not disclosed how many.
West acknowledged that although the SOLiD system is not fully commercialized yet, “as a practical matter, we [have been] competing against SOLiD now for some time,” as customers have had access to information and data sets from both systems.
Even after selling almost 100 sequencers this year, Illumina is confident that manufacturing will not become a bottleneck for its sequencing business. The output of its manufacturing group “has been doubling almost on a quarterly basis for the better part of the year now,” West said. “Although we see it as an important challenge and a lot of work to be done, we don’t see that as a limitation.”
On the Offensive
When talking about the advantages of his company’s sequencing platform, West took a stab at his competitors’ technologies, mainly 454 Life Sciences’ and Applied Biosystems’.
For example, he pointed out that Illumina’s sequencer can work with as little as 100 nanograms of starting DNA, less than competing technologies. The ability to use small amounts of DNA is important in the analysis of cancer samples, West said, “where you just don’t have a lot of material to start with.” He called cancer sequencing “one of the main application markets for next-gen sequencing.”
According to ABI, its SOLiD system uses between 10 and 30 micrograms of DNA, depending on the library. The company said in May that it was working on reducing this amount (see In Sequence 5/29/2007).
West also praised Illumina’s sample amplification technology, which is automated on the company’s cluster station and takes up to two days to generate clusters for sequencing. He also claimed that emulsion PCR, the amplification method used by both 454 and ABI’s SOLiD, is a “complicated and slow process” that “many of the customers that we work with say they absolutely want to get away from.”
According to 454, amplifying a DNA sample by emulsion PCR for its instrument takes about half a day.
In addition, West claimed that the Genome Analyzer, which completes a run in about three days and currently generates about a gigabase of data per run, is faster than some competitors’ tools, which he claimed take up to 10 days for a single run.
454 Life Sciences completes a run in around eight hours, and ABI said in June that its instrument needs three days for a fragment library run and six days for a mate pair library run.
West also pointed to a number of technical improvements to Illumina’s sequencing platform that are currently in the works.
In terms of read length, Illumina currently supports 35-base-pair reads, but the company has already shown that it can go up to 50 base pairs. “Although we don’t push to support this yet, we have lots of our customers who are already leaping ahead and doing it,” West said. Some of these customers have found that “as high as 70 percent of the reads” are “perfect up to 50 bases,” he said.
Illumina plans to combine these 50-base-pair reads with paired-end sequencing, he added, while “every other company that’s in this space is limited by what’s called a ditag sample prep, which limits them to 25 bases [at either end],” he said. “That’s a fairly fundamental limitation built into those technologies.”
However, 454 said earlier this year that it is working on a nebulization-based paired-end method that would yield longer reads on either end (see In Sequence 3/6/2007).
“As we go into next year, we would expect something like 80 or 90 percent of our applications to be done on paired ends.”
Paired-end sequencing, which Illumina is currently beta-testing, will soon become dominant over single-read sequencing, he predicted. “As we go into next year, we would expect something like 80 or 90 percent of our applications to be done on paired ends.”
Paired-end reads, longer reads, and an increased number of imaging columns per channel, another improvement currently in beta-testing, will also help to drive up the output of the platform. West predicted that in the “very near term” each run will produce up to four gigabases of data instead of the one gigabase generated today.
Midterm, meaning over the next year or so, the company expects to increase the cluster density, which “we still have not driven to its limits,” according to West. It is also working to increase the size of its flow cells.
Longer term, or over the next several years, West said, the company is working on increasing the scan speed and improving its chemistry.
These improvements will help customers take on new sequencing applications, he said. Already, customers are trying out new applications on their own even before Illumina comes out with a protocol or kit. West mentioned ChIP sequencing as an example, which led to the first published papers using the Genome Analyzer (see In Sequence 6/12/2006). That application, he said, “has led to a lot of growth in demand for our platform.”
In 2008, as the output of the platform increases, he said Illumina expects new applications to appear that “work at the level of the entire genome,” including whole-genome sequencing and whole-transcriptome sequencing.
“In particular, with the advent of some of the assemblers that are being developed at a number of academic bioinformatics centers, this is a technology now that will be taken into large-genome de novo sequencing,” he said.
Illumina is also betting on synergies between large-scale genotyping and sequencing. The company “already sees numerous customer accounts where the customers in fact are performing both genome-scale genotyping and genome-scale sequencing,” West said.
One customer who presented at Illumina’s meeting last week supported this notion. David Duggan, director of the advanced genomics technology center at the Translational Genomics Research Institute, told In Sequence by e-mail that he discussed “how the [Genome Analyzer] is likely to change the experimental design of both candidate gene studies and follow-up studies to the many genome-wide scans being performed at TGen.”
Overall, West said Illumina is confident that it will continue to grow its sequencing business in the coming years, both by gaining market share and by expanding the market with new applications.
Funding for sequencing is already available, West said. “We have not had to develop a new concept and get scientific groups to realize that this is an area that should be funded,“ he said. “We are moving into a sequencing market that is already spending $1 billion per year.”