by
(
Abstract
Craig Venter reaches for the stars yet again with the
creation of Celera Genomics, and a vow to sequence the human genome cheaper
and faster than the National Institutes of Health.
Wednesday is hot lunch day at Celera Genomics. Or so says the security guard (one of three working reception) who escorts me to the kitchen for a snack and a Diet Coke while I await the arrival of J. Craig Venter, the company's director.
Celera's new headquarters is nestled between Pepco and the local cable TV provider in a cluster of corporate parks in Rockville, Maryland. The company formed in May, when Venter joined forces with Perkin-Elmer, the biotech equipment manufacturer, and announced his intention to sequence the human genome by 2001 - four years sooner than the government-backed Human Genome Project originally promised to finish the job.
The building's temporary reception area is quiet - so quiet I can hear the room lights hum. Perhaps it's just the calm after the storm. Venter's declaration that he could churn out the human genome sequence faster and cheaper than the National Institutes of Health - at a cost of $200 to $300 million dollars compared to their $3 billion budget - made headlines and sparked a heated debate in the research community. Could Venter do it? And if he did, would it render the 15-year government effort redundant? At a congressional subcommittee hearing raised to address such questions, Maynard Olson, director of the genome sequencing project at the University of Washington Genome Center in Seattle, used the phrase "catastrophic failure" to describe the likely outcome of Celera's plan. As Venter recalls, Olson made this prediction not once, not twice, but three times.
Still, Venter remains cavalier in the face of such criticism. "I don't take it personally," he shrugs, digging into leftover Chinese food with a white plastic fork. "We just don't have a federal research system that can accept innovative new approaches. But I've had to work outside the box before."
A TIGR Is Born
Indeed, this is not the first time Venter has circumvented the scientific establishment to accomplish his goals. For anyone who has not read one of the recent profiles - including an 8,000-word cover story in the Sunday New York Times Magazine - Venter is something of a scientific maverick with a history of thinking big and, generally speaking, getting big things done.
Ten years ago, Venter was working at NIH, where his lab served as one of the test sites for the first generation of automated sequencers. Six months after starting with the new machines, Venter's group had sequenced a pair of neurotransmitter receptor genes - quite an accomplishment in those days. But Venter wanted to pick up the pace, which is why he turned to sequencing expressed tag sequences (ESTs). Working with ESTs, which are derived from cDNA libraries that are generated from mRNAs, allows one to focus on gene-rich regions of DNA. Within two years, Venter and his colleagues had found pieces of some 7,000 genes.
Bolstered by his success, Venter applied for a $10 million grant to do large-scale EST sequencing in humans. His application was rejected, and in 1992 Venter left NIH to form The Institute of Genomic Research - TIGR for short. But 35,000 human ESTs later, Venter still was not satisfied. So he set his sights on decoding an entire genome, this time using a technique called shotgun sequencing. At the suggestion of Nobel laureate Hamilton Smith, Venter chose to target Haemophilus influenzae, a bacterium that causes ear and respiratory infections in children. Again, he applied to NIH for funding. By the time this grant was rejected, Venter and Smith had already completed 90 percent of the Haemophilus sequence.
Since the Haemophilus sequence was published in Science in 1995, Venter and his TIGR team have sequenced the complete genomes of seven different organisms, including Methanococcus jannaschii, a deep-sea bacterium, Helicobacter pylori, a microbe that causes ulcers, and Borrelia burgdorferi, the spirochete responsible for Lyme disease. They're also wrapping up the sequence of Mycobacterium tuberculosis, in collaboration with Barry Bloom and Bill Jacobs at the Albert Einstein College of Medicine in New York City. The TIGR researchers have also submitted the sequence of one chromosome from the malaria parasite Plasmodium for publication.
If the shotgun technique worked for these beasties, why not for humans? One problem relates to size. The human genome contains about 3 billion bases compared to a modest 2 million for Haemophilus. Using shotgun techniques to tackle a genome as large and complex as that which makes us human would require some major improvements in sequencing technology. That's where Perkin-Elmer comes in.
The New Gene Machine
"This time they approached me," says Venter. Mike Hunkapiller, president of the Applied Biosystems division of Perkin-Elmer, called to describe their new machine. "He asked if I would be interested in sequencing the whole genome," says Venter, who remained skeptical until he met the ABS 3700. "I knew instantaneously it was doable."
The new sequencer operates on the same basic principle as the older models: it tags DNA fragments with fluorescent dyes, separates them, and then the detector reads them. The resulting sequences are fed into a computer program that assembles them into the correct order. The ABS 3700, the first fully automated sequencing machine, reduces the manpower needed to run a set of samples from eight hours to 15 minutes per day. According to Venter, a technician need only "check to see that the machine has enough reagents and then make sure nobody kicks out the plug."
Venter hopes Celera will have 230 of these machines up and running by April. He anticipates generating 1 million base pairs of sequence each day, accomplishing in one week what TIGR needed one year to finish.
Of course the proof of the pudding is in the sequence. And Venter is the first to admit the enterprise may fail. "This is a grand experiment," he says. "Nobody knows what the outcome will be." Critics involved with the Human Genome Project warn that Venter's sequence will be filled with holes. He says theirs will, too. "They have to find something wrong with what we're proposing," notes Venter. "Otherwise what's the justification for keeping them going?"
When Is a Race Not a Race?
"This is not a race," says Francis Collins, director of the NIH's Human Genome Project. And Venter readily agrees. But somehow Venter's immodest proposal - followed by NIH's September 14 promise to accelerate their efforts to finish a "rough draft" of the genome by 2001, and the entire sequence by 2003 - certainly makes it seem like one. "OK, is it rewarding and exciting to be first?" asks Venter. "Absolutely." But that's not his sole objective. "The goal is to get the human genome done quickly," he says."We have the technology; it would be wrong not to do it."
But before jumping into the human genome, Venter plans to test his method on the 120-megabase Drosophila genome. Not only is the fruit fly an important model organism for understanding human genetics, but researchers have already gone some distance in mapping its genes. Such a map allows scientists to place their sequenced fragments in the correct order.
By skipping the mapping step - for both flies and humans - the Celera scientists can save some time. Venter predicts he and his colleagues can sequence and assemble the Drosophila genome in a month or two. And though Collins worries that the Celera team may have trouble ordering its human sequences properly, Venter doesn't anticipate a problem. He and his colleagues can always use the human maps painstakingly constructed by the NIH-funded researchers to help assemble their sequences.
Whose Genome Is It, Anyway?
Questions of data release may be the scientific community's biggest concern. How quickly will Venter and his Celera colleagues make sequences available to the public? While government-funded researchers post their sequence data on a daily basis, Venter plans to release his data every three months.
And what sort of data will the Celera project make available? According to Collins, raw sequences - strings of As, Ts, Gs, and Cs - are useful, but they don't provide information about genetic variation. And that's where the big money is likely to be. Venter and his colleagues also plan to generate and market a database of human polymorphisms - small genetic differences between individuals that might relate to a person's susceptibility to disease or responsiveness to drug treatments. Collins would like to see that information in the public domain. "It's part of our responsibility as leaders in genomics," he says.
Post-Genomic Era
In the long run, Venter also wants to get the data out there so the real fun can begin. "We don't sequence DNA because we like sequencing," he says."It's the only way to get the data we need to describe the complete biology of a species based on its genes."
Whether Celera and the NIH effort will find a way to work together to complete the human sequence remains to be seen. According to Venter, finishing the sequence, whether it takes three years or five, is just the beginning. "At the end of the next century, I think people will still be trying to piece together the big picture: understanding how genes function as a blueprint of life."
Until then, Venter is just doing what he can. "I don't necessarily like the idea of becoming the poster child for genomics. And I'm as tired of seeing my name in print as I'm sure a lot of people are," he says. "But if it helps the public understand what we're trying to do, I think it's useful."
So Venter isn't bucking to become the Bill Gates of the human genome? "I'm not sure I know what that means," laughs the scientist, finishing off the egg roll he's saved for last. "But somehow I don't think it's meant as a compliment."
Karen Hopkin, a freelance writer and editor, received her Ph.D. in biochemistry from the Albert Einstein College of Medicine in 1992. She is the creator of the Studmuffins of Science Calendar.
Endlinks
TIGR Database - provides access to microbial and parasitic sequence data and human ESTs.
Private Venture Galvanizes Public Effort on Human Genome Project - more details about the Venter and Perkin-Elmer initiative from the June 24, 1998 issue of the Journal of the American Medical Association.
Primer on Molecular Genetics - although the sequencing information is out of date, this site introduces mapping strategies of the Human Genome Project.
Kits to DYE for: A Profile of Sequencing Kits for Automated DNA Sequencers - a recent comparison of products from eight companies. From The Scientist's LabConsumer column.
Web sites mentioned in this column:
- Celera Genomics
- Perkin-Elmer
- National Institutes of Health
- The Institute of Genomic Research
- Science
- Albert Einstein College of Medicine