New Sequencing Center Aids Intramural Science
By Rich McManus
On the Front Page...
In a low-rise biotech beehive just off I-270 in Gaithersburg, the new NIH Intramural Sequencing Center -- a 14-institute consortium that is one of perhaps a half dozen nonprivate centers in the United States dedicated to large-scale sequencing -- is quietly parsing all the A's, T's, C's and G's that make up stretches of human and animal DNA of interest to bench scientists. Showing up on the computer screen as rosaries of red, blue, green and yellow, the nucleotides are exposed and decoded so that scientists such as NIDCD's Dr. Thomas Friedman can make discoveries such as appeared in the May 29 issue of Science: mutations in an unconventional myosin gene are another cause of hereditary deafness.
The very banality of the architecture at 5 Research Court, where NISC is temporarily located before moving later this summer to quarters roughly twice as large, announces that things of great importance are going on. Tucked amid a Quality Inn for executives on the run and neighboring low-rises with discreet signs designed to give away little about what's happening within, 5 Research Court is the postmodern equivalent of the senates and cathedrals of yesterday. Unlike those antediluvian temples of consequence, however, NISC and its peers announce a world in a hurry, the frothy leading edge of a breaking wave. The preoccupations of hot-field progress do not permit ornament.
On the morning after a private venture led by an ex-NIH scientist announced that it would scoop the federal effort to sequence the entire human genome, NISC was an optimistic, hungry, undeterred place.
"We think that [private efforts to sequence genomes] will expand interest in our program," said Dr. Jeff Touchman, director of the sequence production group at NISC and a former postdoc in the NHGRI laboratory where the center got started. "It's going to create more of an appetite for the kind of work we do. In the post-human sequence era, a lot of resequencing will be performed, plus studies of the genomes of model organisms. This will add value to the work in human genomics. There are numerous human cell types whose genes are still unsurveyed, despite the efforts of TIGR [The Institute for Genome Research, an NIH-bred biotech group] and other centers, and the slope for technology development is sharply upward in this field."
Opened last September, NISC "is the brainchild of Dr. Eric Green (chief of NHGRI's Genome Technology Branch)," Touchman said. "He and I were approached by a number of investigators on campus to do sequencing for them. Eric had two sequencing instruments at the time, and began to think there may be a place at NIH for a high-throughput sequencing center, capable of producing multiple megabases per year.
"Many institutes have their own sequencing machines, and analyze maybe 100 to 200 reactions per week," he observed. NISC currently boasts six ABI 377 sequencing instruments and routinely uses them to analyze 3,000 reactions per week, or upwards of 6 megabases of sequence a year. A new instrument expected this summer -- a capillary electrophoresis machine -- "will increase throughput even farther," Touchman predicted. "Our goal at NISC is to do something that individual labs on their own can't do, by taking advantage of the economy of scale gained by our dedicated, larger operation."
NISC offers three basic services, said Touchman, a 4-year NIH veteran who was busy with genomic mapping of human chromosome 7 before branching out into sequencing:
Finished Genomic Sequencing -- "We take a large clone and sequence the whole thing to high accuracy. This could be upwards of 100 kilobases of DNA or more," he explained. NISC has completed half a megabase of "finished" sequence in the 8 months since it opened.
EST (Expressed-Sequence Tag) Sequencing -- "Basically, gene surveys. It's a very common intramural request," Touchman said. "What genes are expressed in a given tissue or organism? We can sequence a large sample of the RNA to begin to find out. That can lead to all sorts of interesting downstream biology." NISC has generated 12,000 EST sequences so far.
Sample Sequencing -- Also known as low-redundancy genomic sequencing, it is "a very powerful method for positionally cloning a gene. Specifically, this involves generating a collection of random sequences across a broad genomic interval. We've sampled over a megabase to date."
Just 10 people comprise NISC at the moment, though it hopes to add more employees: 7 work on the DNA sequencing production team, and 3 are computational biologists -- specifically trained interpreters of the strewn alphabet of nucleotides, guided by Gerry Bouffard and Stephen Beckstrom-Sternberg. "They make sense of the sequences we produce," noted Touchman. "The sequence data needs lots of massaging at the end for it to be meaningful." Most hail from NHGRI. And all face sore temptation to run down the street to the next tree-shrouded low-rise where private ventures such as TIGR and Human Genome Sciences -- outfits roughly 8-10 times the size of NISC -- offer higher salaries.
NISC is proud of its record of training young people, he added. "Lots of folks from NIH learn about automated sequencing from us. Pre-IRTA fellows get high-throughput training with NISC."
While the data generated by NISC belongs to whomever pays for it -- it's a fee-for-service operation -- Touchman says about half its clients hold on to the data until publication of a paper, and another half send the sequences directly to GenBank, an NIH-sponsored repository. "We have no policy on the immediate release of data, but encourage submission to GenBank as quickly as possible," he noted.
So far, NISC has completed 25 large-scale projects for intramural scientists. There is currently a 4- to 6-week queue, but the center would like to bump that up to 8. "We're a little bit underutilized considering all the projects we suspect are out there," Touchman said. "The rank and file scientist at NIH needs to know we're here."
The minimum project size is 500 sequences, which can take only a few days to read. NISC can process some 3,000 reads a week, or 600 per day. A 15-member scientific review committee evaluates all incoming applications, most of which succeed without amendment, reports Touchman. "It's basically like a grant review," he observes, "a short 2-page research summary. We're averaging about two proposals a month, but would like to double that."
The typical NISC project takes 3-4 weeks, and multiple projects occur simultaneously. Output will increase when NISC moves to a new building in July. "We'll have twice as much space in the new facility, with NCI also having their own sequencing center next door," Touchman said.
Up to Top