ROCKVILLE, Md.--The highly publicized May 9 announcement by Craig Venter of the Institute for Genomic Research (TIGR) here and Michael Hunkapiller of the Applied Biosystems division of Perkin-Elmer of their plan to utilize a new Applied Biosystems instrument to sequence the entire human genome has spurred speculation about the project's potential effects on the bioinformatics industry. The scientists announced that they will create a company financed by Perkin-Elmer that will be in full production mode by April 1, 1999, and will complete the sequence in just three years for under $200 million.
While some experts who spoke with BioInform were skeptical of the new company's chances to succeed at that rate or budget, others expressed concerns about public access to the data they believe Venter and his company will uncover. "Will this speed up the development of the entire bioinformatics industry or curtail it because the information will be in the hands of a private entity?" one observer wondered.
Just what exactly the new Applied Biosystems instrument is remains something of a mystery. Senior executives of the company refrained from discussing the new product during presentations at pharmaceutical industry conferences in Philadelphia and Baltimore after the announcement. In a press release announcing that the company was "nearing the end of a breakthrough DNA analysis technology" Perkin-Elmer said only that the 3700 DNA Analyzer, to be priced at $300,000, "combines proven capillary electrophoresis hardware and separation polymer chemistry with new detection technology." The instrument is designed to enable DNA sequencing and genetic analysis applications requiring tens of thousands of samples produced weekly, the company claimed.
"Whether it's evolutionary or revolutionary and what will enable Perkin-Elmer to accomplish what they say they can is not clear to me," said an industry insider who spoke to BioInform on the condition of anonymity. "I don't know how much of it is related to the quality and power of the instrumentation that Perkin-Elmer is providing versus how much is due to new sample prep techniques that they or Craig Venter may have developed."
TIGR bioinformatics director Anthony Kerlavage, whose role in the new venture is still to be determined, said the company will "marry the expertise and technologies developed at TIGR with a real leap forward in sequencing technology." Kerlavage described the instrument as an automated loading, capillary-based system and an evolved technology combined with many new features. "It's a number of things converging," he said. "Applied Biosystems has a single capillary sequencing instrument. The new one can handle 96 samples instead of one." New built-in features also allow automatic loading and incorporate evolved sequencing chemistries, he said.
Kerlavage explained, "The big advantage is that it can process 1,000 samples a day, requiring only 15 minutes of operator time per 24 hours of operation." Instruments used at TIGR now demand about 8 hours of hands-on time per day and two workshifts to perform three runs a day, including a long overnight run, he said. The new company intends to do 10 runs per day of 96 samples on each of the more than 200 instruments it plans to install, he said.
"This will create a unique opportunity to look at the complete human genome in the same way we've become accustomed to with microbials," Kerlavage said. "From an informatics and functional biology perspective, this is what we've been looking forward to."
Some of Kerlavage's bioinformatics peers are not as enthusiastic. "My concern from an informatics point of view is about access to the data," said a government bioinformaticist. He cited a media report that quoted Venter remarking that he planned to avail the public of all but the 100-300 genetic sequences that look commercially promising.
"The 300 most valuable human gene sequences will almost certainly be worth an enormous fortune," the government employee commented.
But Kerlavage said all data will be made available through GenBank on a quarterly basis and that research will be focused on the downstream characterization of those 100-300 valuable genes after sequencing. "The sequence data will still be public. The patents will be focused on the particular area of research of the gene or protein," he said.
He said the new company will have three profit drivers: revenue will be earned from patent protection and licensing, from marketing the comprehensive human genomic database to pharmaceutical and biotech companies, and from the demand for new systems and tools for looking at single nucleotide polymorphisms (SNP's). The yet-to-be-named company also plans to generate assay systems to validate those SNP markers, which would be sold by Perkin-Elmer, he said.
The announcement has raised other debates, too. One is over the value of full-length sequencing. Venture capitalist Beckie Robertson of Institutional Venture Partners in Menlo Park, Calif., said, "It's been clear that the underlying assumption for some time has been that express sequence tags and polymorphisms were all we really needed to understand and that the exercise of sequencing the full human genome was more of an academic one. Perkin-Elmer and TIGR have precipitated a debate about whether in fact that will be enough."
Another hotly debated topic raised by the new project is whether the government should drop its $3 billion human genome sequencing project. The Venter venture, slated to complete the genome two years ahead of the government schedule, would make the national project redundant, many experts have suggested. But other disagree. "I object to rhetoric that suggests the government should get out of the human genome sequencing business," said a Washington-based bioinformaticist. "It would be a shame if the NIH-sponsored labs stopped doing human sequencing work, because there is a the risk of Venter failing, and there are intellectual property implications of a taxpayer-funded human genome sequence."
Implications for informatics
The news also indicates that the post-genomic era is approaching at a faster rate than has been expected. But what that means for bioinformatics businesses isn't clear.
Kerlavage predicted that the project will create plenty of new opportunities. "This will pose large bioinformatics challenges," he said. "We're proposing a 3 gigabase genome that will create many technical problems. We're making improvements to handle the data. But it will create problems downstream, too."
Bioinformatics tools in use today will not scale well to the new quantities of data, Kerlavage predicted. "We will have to completely rescale the pipeline for analysis of these data," he said.
One immediate effect of the news was apparent at Palo Alto, Calif., human genome database provider Incyte Pharmaceuticals. The day after the New York Times broke the news of the Venter/Hunkapiller collaboration on the front page of its Sunday, May 10 edition, analysts lowered their ratings on Incyte stock, saying the new deal could accelerate the devaluation of commercial databases such as Incyte's. But others who follow the industry told BioInform those reactions were premature.
"It's clear that Perkin-Elmer has Incyte in their bull's-eye," said IVP's Sam Colella. "There's no question there will be an impact on Incyte. But they've known that's coming. If you look at their product line, they've got a lot more content than the human database."
"I think we're all having a hard time looking two or three moves ahead in this game," Robertson added. "The underlying assumption is that when the first human genome is sequenced the game is over. Yet it's hard for me to believe that what it isn't going to spawn is a whole new set of data that's going to be extraordinarily important."
"It remains to be seen what the effect will be on database providers such as Incyte," Kerlavage said. "Various companies are providing value-added databases. The new company will have a complete set with annotation. We will have more available than any other database. That's what would drive a customer to us," he said.
Kerlavage said there's no question TIGR will be getting out of the human genome sequencing business. Instead, the institute--now under the leadership of Venter's wife, Claire Fraser--will focus on microbial and plant sequencing, and will eventually install the new Applied Biosystems equipment too. Kerlavage said a higher level of collaboration between Perkin-Elmer and TIGR is expected. A facility under construction on the TIGR campus here will house part of the new venture's project, and another nearby facility is being sought to house its bioinformatics department and several hundred new hires.