Decoding the Lowly Fruit Fly
Campus Scientists Team with Genetics Company to Speed Up Sequencing
By Robert Sanders, Public Affairs
Gerry Rubin, director of the Berkeley Drosophila Genome Project Group, and J. Craig Venter, founder and president of Celera Genomics Corp. of Rockville, Md., signed a memo of understanding last month pledging to work together to complete the sequencing by the end of 1999. The two geneticists hope that Celera's patented technique, called "shotgun sequencing," will save time and money.
"This should accelerate things quite a lot, allowing us to finish in a year rather than another three years, which was the original plan," Rubin said.
The goal is an important milestone on the road to sequencing the entire human genome, he said.
"Because the fruit fly is more similar to humans than any other animal sequenced so far, its genetic information can be directly related to humans in many cases," he said. As one of the most popular research animals for geneticists, "the animal has provided an extraordinarily deep and broad understanding of the function of genes in disease."
If the experiment works as hoped, the collaboration could save the government DNA sequencing project $10 million or more. The U.S. Human Genome Project and Berkeley's part of it are funded through the National Human Genome Research Institute, a component of the National Institutes of Health, and through the Department of Energy.
DOE's Lawrence Berkeley National Laboratory plays a key role in the fruit fly genome project directed. In December Rubin received another $40 million to complete the sequencing over the next four years.
"This is really a test of how well Celera's method works on larger genomes," Rubin emphasized. To date Venter has wielded the "whole-genome shotgun method" to sequence the much smaller genomes of nearly a dozen bacteria, including Hemophilus influenzae, the cause of various respiratory and neurologic infections.
Though Venter has bragged that he will sequence the human genome within three years -- far ahead of the federally funded Human Genome Project, which aims for completion in 2005 -- in fact his technique is unproven on genomes as complex as that of fruit flies, mice or humans.
"It is also a test of whether a public-private collaboration can expedite and lower the cost of the generation and use of genomic information that will ultimately benefit agriculture, and prevention, diagnosis and treatment of important human illnesses," according to the memorandum of understanding.
Speaking Jan. 25 at the fourth annual James D. Watson lecture in Washington, D.C., Francis Collins, director of NHGRI, said, "This agreement marks the beginning of a productive collaboration between the public and private genome sequencing programs ... (which) will also provide an important pilot for the development of a similar partnership effort to obtain the human sequence."
The fruit fly sequencing project was first funded in 1992 by the Human Genome Project, with Berkeley as the base of operations and Rubin as head of a consortium of universities and federal labs. The first three years were spent mapping the various cloned snippets of fly DNA, while sequencing began about three years ago. The plan was to sequence the fruit fly's 150 million base-pair genome by 2001. Berkeley researchers have already produced about 20 percent of the fly's DNA sequence in high-quality form.
Under the new arrangement, the Berkeley group will increase their activities in parceling up the fly genome for sequencing by copying and mapping overlapping DNA pieces that span the fly's genome.
Meanwhile, researchers at Celera, a business unit of The Perkin-Elmer Corp., will try out their "shotgun" sequencing strategy. The whole-genome shotgun method will produce sequences from many small, random DNA fragments. By combining sequences with map information and additional sequences provided by the Berkeley group, the team aims to assemble the data into long stretches of correctly ordered, continuous genomic sequence.
Both groups will work to fill in any remaining gaps and ensure the sequence meets quality standards. Celera will deposit the data it produces in batches in the public-domain database, GenBank. The complete set of shotgun data is expected to be in GenBank by the end of 1999.