By Mark D. Uehling
November 15, 2002 | It's 2008, and your only daughter is turning 13. It's her last visit with the family pediatrician, who proudly presents her with a DVD. It's her own personal genome, annotated and sequenced down to the last A, C, T, and G. Maybe it will contain auspicious tidings: a gene for longevity, perhaps. Maybe not …
Either way, the news in Boston at The Institute for Genome Research's (TIGR) 14th International Genome Sequencing and Analysis Conference in early October is that the equipment to give everyone a personal genome for the price of a cheap laptop is closer than even conference organizers suspected.
On the opening evening of the meeting, TIGR founder J. Craig Venter hosted presentations from six scientists, each of which poses a potential threat to Venter's former backer and the meeting’s sponsor, Applied Biosystems Inc. All but one presenter believed a $1,000 genome -- or more accurately, genotype -- is feasible.
For most of the panelists, the issue is not if but when -- and whether any new concoction of chemistry, optics, fluidics, and software could be profitably combined into a single benchtop instrument that could seriously challenge current sequencing machines costing several hundreds of thousands of dollars.
Beyond the Sequence
"This is inexpensive, off-the-shelf technology," said George Church of Harvard Medical School. He displayed just three main components: an MJ Research in-situ cycler, an automated histology slide rack, and a GenePix microarray scanner. "We can go from cells to DNA," Church said. He tabulated the cost of a genome at $710, including $150 for electricity.
Michael Weiner, vice president of molecular sciences for 454 Corp., a CuraGen spinoff, debuted a prototype microfluidic slide about the size of a stick of gum, containing 1 million tiny wells on the plate. For now, 454 is just sequencing viruses and bacteria, Weiner said. But the company will follow Venter's lead and try its hand at more complex organisms.
That work is forcing 454 to soberly face the IT requirements of any next-generation DNA sequencer. For now they are running Linux clusters, but gathering the information for an individual genome "is not for the faint of heart," Weiner said. "It takes a lot of computing power."
An Uncut Helix?
Perhaps the most brazenly confident presentation was by Eugene Chan, the founder of U.S. Genomics Inc. Touting 36 company patents and instruments available now, Chan explained the company's platform to unspool the DNA spiral and pass a linear molecule past a laser scanner. "The DNA is uncoiled, unfurled, and read past a fixed reader just like a movie reel spools over a projector," Chan said.
Because he can already scan DNA in much longer stretches than is common in the industry -- as many as 200,000 base pairs at a time -- Chan was the most optimistic about lowering the cost of sequencing a human being. "The molecules move by at 30 million bases of DNA per minute," he said. "In a 40-minute time frame, you get about 3 billion base pairs of DNA going through the system."
If the whole system can be transferred to a chip, Chan said, the time could drop even more, to less than five minutes. "Our goal," he said, "is to be able to read your genome instantaneously." The catch is that the company will have to develop the ability to capture a terabyte of information every few seconds, Chan said. That bioinformatics software is under development.
Britain-based Solexa Ltd. also hopes to combine nanotechnology and genomics. Chief technology officer Tony Smith notes the company’s single molecule arrays consist of 100 million randomly arranged oligonucleotides within a square centimeter. Each successively added nucleotide is scanned and read sequentially using fluorescence. Rather than sequencing individual genomes from scratch, Solexa will compare 25-base DNA fragments to the published human genome sequence.
“We’re able to target all the differences between the sample and the human genome reference,” Smith said, adding that Solexa’s so-called “one-pot” approach “is a revolutionary improvement in throughput and cost.”
Trevor Hawkins of Amersham Biosciences cautioned that "there is technology that will get us down to a $30,000 genome in the next few years, but to get us to the next level will take a new technology." By his estimate, half the money would still be used for reagents. The cost will have to be driven almost straight down to zero, Hawkins suggested.
Houston's Heavy Hitter
Susan Hardin is president and CEO of VisiGen Biotechnologies Inc., the first tenant at a new Houston biotech incubator park. VisiGen's approach, she said, could produce a human genome in an hour or two for $1,000 (or as she would market it, $995). A VisiGen instrument could be two to four years away; the company has grants from the National Institutes of Health and the Defense Advanced Research Projects Agency.
The key for VisiGen will be tweaking DNA polymerase to identify individual nucleotides during DNA replication. Like Solexa, Hardin’s company is hoping to avoid the time-consuming duplication of DNA. She hopes to eliminate several expensive, time-consuming chemical steps in traditional genotyping and read the DNA directly. “There is no reaction assembly followed by cleanup,” Hardin said.
The other appeal of her approach, she said, is the promise of direct access to genetic data. "It's an incredibly information-rich way to design an assay system,” Hardin said. “You're getting into the reaction and visualizing the bases. It's real-time data acquisition. You can work in massively parallel processing."
At the end of the evening, Venter himself professed to be, if not entirely convinced, certainly "enthused -- and more optimistic than I was earlier in the day."