In late 1997, an ambitious idea occurred to technology guru Michael W. Hunkapiller of Perkin-Elmer Corp. Hunkapiller's team was developing a robotic machine that promised to decipher human genes far faster and more cheaply than any previous system. Why not use the new device, Hunkapiller wondered, to tackle one of the biggest prizes in all of biology--successfully deciphering the entire human genetic code? He brought his idea to gene sleuth extraordinaire J. Craig Venter, president of the nonprofit Institute for Genomic Research in Rockville, Md.
The result, announced on May 9, is a still unnamed company that will decipher what one "might describe as the full Monty--the entire genome," says Venter. With some 230 of the new $300,000 Perkin-Elmer machines running around the clock, Venter and colleague Mark Adams figure they can break the 3 billion individual units of human DNA--the genome--into pieces and decode a staggering 100 million individual units a day. They plan to finish the genetic code in three years, at a total cost of about $200 million--with Perkin-Elmer picking up the tab. That is a fraction of what the federal government is spending to complete the task--and Venter vows to finish four years sooner.
What's more, Venter and Perkin-Elmer will give away the entire human DNA sequence, just as the government plans to do. We agreed it would be morally wrong to hold the data hostage," says Venter. The gamble for Perkin-Elmer--a pioneer in gene sequencing--is that it can make money by selling information about what the sequence means, as well as finding new genes for developing medical therapies.
"BLACK EYE." The announcement sent shock waves through the red-hot field of gene-mining. This discipline, called genomics, is already populated by dozens of companies (table, page 72) and academic labs seeking to understand and profit from DNA's secrets. Companies such as Human Genome Sciences Inc. (HGS) and Incyte Pharmaceuticals Inc. have already made millions selling access to their private stashes of gene sequences. But the new company is a formidable competitor--"a 1,000-pound gorilla," says analyst Elizabeth Silverman of BancAmerica Robertson Stevens. Adds Randal W. Scott, president of Incyte in Palo Alto, Calif: "This puts a new competitor into play." And the idea that a private company can soundly beat the existing taxpayer-funded effort to the prize "is a tremendous black eye for the government," says William A. Haseltine, CEO of HGS. "They will lose the race to the genome."
But the venture also raises a host of questions. Does the massive private effort mean that the government's Human Genome Project should redirect its efforts? And will Perkin-Elmer actually be able to make money from its radically different business plan?
On the science, few are betting against Venter. "There's no question that the person who can put together an operation like this and make more headway than anyone else is Craig Venter," says Stanford University biochemist and Nobel laureate Paul Berg. Back in the mid-1990s, Venter pioneered a "shotgun" approach to deciphering entire genomes. The idea was to chop the DNA of an organism into pieces, decipher each of them, and then use computers to compare and assemble them in the right order. Using the technique, Venter astounded the scientific world by decoding the first complete genetic sequence of a living organism--a bacterium called Haemophilus influenzae.
Perkin-Elmer's new machines will speed up the process. Its Applied Biosystems Division sold $650 million worth of DNA sequencers and related instruments and services in fiscal 1997. The new tool, available next year, "is an evolution of our current system," says Hunkapiller. Its improved sensitivity and automation will dramatically boost productivity.
DATA FLOOD. Venter is hinting that the government's genome project should shift its focus to, perhaps, sequencing the DNA of animals instead of people. That's not likely. Dr. Francis Collins, head of the National Institutes of Health's genome center, wants more proof that the new company will live up to its promises before he alters his course. And even if Venter succeeds, making sense of the flood of information won't be easy. Only about 3% of human genetic material is actual genes. Some of the remaining 97% of the DNA turns genes on and off, and scientists think that much of the rest is meaningless junk. Part of Venter's job will be to figure out what's what, and that could be tough. "The genes jump right out at you in microbial sequences," says Richard K. Wilson of Washington University's gene-sequencing center. "In humans, it's much more difficult."
Many are confident of Venter's scientific claims, but the business end of this venture is another story. Perkin-Elmer faces an uphill battle convincing the biotech world that this is a money-making idea. "What they're describing is not a commercial venture," says Incyte's Scott. "It's really Craig Venter going after the Nobel prize for sequencing the genome." HGS's Haseltine is also skeptical. "The human genome project has never been a commercial venture," he says. "This is more in the tradition of the Mellons and Carnegies"--funding a project that promises mainly to push back the bounds of knowledge.
Perkin-Elmer execs insist that their proposal has been misunderstood. "People still don't see how, if we give away the data, we will make money," sighs CEO Tony L. White, as he patiently explains the plan. Stanford's Berg says that "the big game is how to make use of the information," and that's the information White plans to sell. Rival Incyte is already an old hand at this. In fact, one of its products is a repackaging of publicly available data in more usable form, says analyst Mike G. King of Vector Securities International. Haseltine wonders how Perkin-Elmer can do this "better than the rest of the world combined."
Venter and Perkin-Elmer execs retort that the new company will have enough experience and smarts to be a leader in this toughly competitive field. They envision signing up hundreds of thousands of subscribers--both companies and academics--for a database that offers such vital information as which sequences are genes, what the genes do, and how genes can vary from person to person. Such variations, called "polymorphisms," determine whether individuals are susceptible to certain diseases or how well drugs will work. Doctors and pharmaceutical companies can use the information to better diagnose and treat people based on their genetic variations. And companies such as Affymetrix Inc. will benefit, analysts predict. Affymetrix makes gene chips, which can almost instantly spot the presence of thousands of different genes or gene variations.
DRUG DEVELOPMENT. Perkin-Elmer should also benefit. The $1.4 billion company has moved aggressively to acquire companies and new technology, transforming Perkin-Elmer from an instrument maker to one that provides services and information as well. Since White took over in 1995, the company has acquired Tropix, a leader in screening drug candidates, and GenScope, developer of gene expression technology, and forged partnerships with other players. For instance, it teamed up last June with gene-chip developer Hyseq Inc., whose products can be used to search for gene variations.
Venter's and Perkin-Elmer's venture may also profit from new genes that Venter finds. The main current approach for finding genes involves fishing out those that are actually turned on in cells. Venter argues that this tack, which, ironically, he pioneered, misses some of the genome's real gold. That's because some genes may turn on too rarely to be discovered. He estimates that by sequencing the entire genome, he'll find 10,000 to 20,000 new genes. Many will be genes for vital signaling pathways in the body and brain--ideal candidates or targets for drugs. As a result, "these genes will have tremendous value on their own," he says. He expects the new company to pluck out a few hundred of the most promising to patent and use for drug development.
The risks, of course, are high. Haseltine and others think the new company may very well succeed at deciphering the entire human genome. Making money, however, will be harder. Venter knows that, but thinks he'll prove the skeptics wrong within a year. By then, he and his supporters believe, the new tools will prove their worth, and vindicate Venter's hunches once again.