THE COMPANY forged from Perkin-Elmer amid the turmoil was the PE Corporation, which was divided into two pieces, the PE Biosystems Group, the unit that was making the Prism machines, and Celera Genomics, which was using the machines to decode the human DNA. Michael Hunkapiller, who became the president of PE Biosystems, believed that he could sell a lot of machines to everyone, including to the Human Genome Project. Craig Venter’s project would demonstrate how effective the Prism machines were; it was advertising. The deal was that there was a fat profit margin in the chemicals the machines used. The chemicals had a much higher profit margin than the machine; not only that, but the chemicals actually cost far more than the machine over the machine’s lifetime. This was the razor-blade principle: if you put inexpensive razors in people’s hands, you will make money selling blades.
James Watson quietly went to some key members of Congress and persuaded them to spend more money on the public project. At the same time, the leaders of the project announced a radical new game plan: they would produce a “working draft” of the human genome a year
Michael Morgan, of the Wellcome Trust, told me what he thought had happened with the creation of Celera. “From the first press release, Craig saw the public program as something he wanted to denigrate,” Morgan said. “This was our first sign that Celera was setting out to undermine the international effort. What is it that motivates Craig? I think he’s motivated by the same things that drive other scientists—personal ego, a degree of altruism, a desire to push human knowledge forward—but there must be something else that drives the guy. I think Craig has a huge chip on his shoulder that makes him want to be loved. I actually think Craig is desperate to win a Nobel Prize. He also wants to be very, very rich. There is a fundamental incompatibility there.”
One day, I ran into a young player in the Human Genome Project. He believed in the worth and importance of the public project and said that he had turned down a job offer from Celera. He didn’t have any illusions about human nature, or about any of the major players. He said, “Here’s why everyone is so pissed at Craig. The whole project started when James Watson persuaded Congress to give him money for the human genome, and he turned around and gave it to his friends—they’re the heads of centers today. It grew into a lot of money, and then the question was, Who was going to get the Nobel Prize? In the United States, there were seventeen centers in the project, and there was no quality control. It didn’t matter how bad your data was, you just had to produce it, and people weren’t being held accountable for the quality of their product. Then Celera appeared. Because of Celera, the NIH was suddenly forced to consolidate its funding. The NIH and Francis Collins began to dump more than eighty percent of the money into just three centers—Baylor, Washington University, and MIT—and they jacked everybody else. They had to do it, because they had to race Celera, and they couldn’t control too many players. So all but three centers were cut drastically, and some of the labs closed down. Celera was not just threatening their funding but threatening their very lives and everything they had spent years building. It’s kind of sad. Now those people hang around meetings, and the leaders treat them like ‘If you’re really nice, we’ll give you a little piece of the mouse genome.’ That’s the reason so many of them are so angry at Celera. It’s easier for them to go after Craig than to go after Francis Collins and the NIH.”
AT CELERA’S HEADQUARTERS in Rockville, I was shown how human DNA was shotgunned into small pieces when it was sprayed through a hospital nebulizer that cost a dollar fifty. The DNA fragments were then introduced into E. coli bacteria and grown in glass dishes. The bacteria formed brown spots—clones—on the dishes. Each spot had a different fragment of human DNA growing in it. The dishes were carried to a room where three robots sat in glass chambers the size of small bedrooms. Each robot had an arm that moved back and forth rapidly over a dish. Little needles on the arms kept stabbing down and taking up the brown spots. Later, the bits of human DNA in the bacteria would be separated from the bacteria and run through the sequencing machines, producing little bits of human DNA code.
Craig Venter stood watching the robots move. The room smelled faintly like the contents of a human intestine. “This used to be done by hand,” he said. All the human DNA fragments would eventually wind up in the Prism sequencing machines, and what would be left, at the end, was a collection of up to twenty-two million random fragments of sequenced human DNA. Then the river of shattered DNA would come to the supercomputer, and to a computer scientist named Eugene Myers, who with his team would assemble all the broken bits of human code into the more-or-less correct order, producing the full human genome.
GENE MYERS had dark hair and a chiseled, handsome face. He wore glasses, a green half-carat emerald in his left ear, and brown Doc Martens shoes. He also had a ruby and a sapphire that he would wear in his ear, instead of the emerald, depending on his mood. He was sensitive to cold. On the hottest days of summer, Myers wore a yellow Patagonia fleece jacket, and he kept a scarf wrapped around his neck. “My blood’s thin,” he explained to me. He said the scarf was a reference to the DNA of whatever organism he happened to be working on. When I first met Myers, he was keeping himself warm in his fruit-fly scarf. It had a black-and-white zigzag pattern. Later, Myers started wearing his human scarf, which had a green chenille weave of changing stripes. He intended his scarf to make a statement about the warfare between Celera and the public project. “I picked green for my human scarf because I’ve heard that green is a positive, healing color,” he said. “I really want all this bickering to go away.” His office was a cubicle in a sea of cubicles, most of which were stocked with Nerf guns, Stomp Rockets, and plastic Viking helmets. Occasionally, Myers would put “Ride of the Valkyries” on a boom box, and in a loud voice he would declare war. Nerf battles swept through Celera whenever the tension rose. Myers fielded a compound double-action Nerf Lock ’n Load Blaster equipped with a Hyper Sight. “Last week we slaughtered the chromosome team,” he told me.
IN THE FALL, Venter announced that Celera had completed the sequencing of the fruit fly’s DNA and had begun to run human DNA through its sequencing machines—there were now three hundred of them crammed into Building One in Rockville. The Command Center was up and running, and from then on Celera operated in high- speed mode. One day that fall, I talked with the company’s information expert, a stocky man named Marshall Peterson. He took me to the computer room, in Building Two. To get into the room, Peterson punched in a security code and then placed his hand on a sensor, which read the unique pattern of his palm. There was a clack of bolts sliding back. We pushed through the door.
A chill of cold air washed over us, and we entered a room filled with racks of computers that were wired together. “We have fifty-five miles of fiber-optic cables running through this building,” Peterson said. Workmen standing on ladders were installing many more cables in the ceiling. “The disk storage in this room is five times the size of the Library of Congress. We’re getting more storage all the time. We need it.”
He took me to the Command Center, where a couple of people were hanging around consoles. A big screen on the wall showed CNN Headline News. “I’ve got a full-time hacker working for me to prevent security breaches,” Peterson said. “We’re getting feelers over the Internet all the time—people trying to break into our system.” Celera would be dealing with potentially valuable information about the genes of all kinds of organisms. Peterson thought that some of what he called feelers—subtle hacks and unfriendly probes—had been emanating from Celera’s competitors. He said he could never prove it, though. Lately, the probes had been coming from computers in Japan. He thought it was American hackers co-opting the Japanese machines over the Internet.
By October 20, forty days after Celera started running human DNA through its machines, the company announced that it had sequenced 1.2 billion letters of human code. The letters came in small chunks from all over the genome. Six days later, Venter announced that Celera had filed provisional patent applications for sixty-five hundred human genes. The applications were for placeholder patents. The company hoped to figure out later which of the genes would be worth patenting in earnest.
A gene patent gives its holder the right to make commercial products and drugs derived from the gene for a period of seventeen years. Pharmaceutical companies argue that patents are necessary, because without them businesses would never invest the hundreds of millions of dollars needed to develop a new drug and get it through the licensing process of the Food and Drug Administration. (“If you have a disease, you’d better hope someone patents the gene for it,” Venter said to me.) On the other hand, parceling out genes to various private companies could lead to what Francis Collins referred to as the “Balkanization of the human genome,” a paralyzing situation that might limit researchers’ access to genes.
Venter insisted that Celera was an information company and that patenting genes was not its main goal. He
