by Chip Walter
Some DNA proteins might deliver the building blocks of an enzyme that helps digest popcorn, or a hormone that fires sexual desire, or a series of molecular interactions that spark a neuron. Or they could form a gene that goes haywire and creates a disease. It was known, even in the 1980s, that genes often serve multiple purposes in combination with other genes—but what those combinations were was almost entirely unknown. Those that were known had been painstakingly hunted down by researchers who squinted at x-rayed images of fuzzy columns on a gel that then revealed the strings of each tagged base pair of DNA by their color: an A (green) or a C (magenta) or a G (blue) or a T (red). It was brutal work.
Scientists who supported what came to be called the Human Genome Project, or HGP, wanted to get to the bottom of what all those genes communicated. In them, they believed, lay the precise blueprint of all humans at the same time it delivered the personal recipe for each human.
Biologists were just beginning to use the growing power of computers to unravel the whole mess, but the process was still slow. In the mid-1980s, scientists at Stanford ran a computer simulation of a single cell dividing. It required half a gigabyte of data and took 10 hours to generate. And that was the gold standard! The experts were quite certain the whole undertaking would require an unprecedented leap in computing power—because ultimately, that was where the information about the genome would have to reside. It was the only way.
Despite these colossal difficulties, Walter Gilbert, a Nobel laureate from Harvard, had stood up at the Cold Spring meeting and argued that the human genome was nothing less than “the holy grail” of biology. True, it would be expensive and massively difficult—but doable if science and government threw “thirty thousand person years” and three billion dollars at the job. That came down to a dollar per DNA base pair. Even for the federal government, that was a lot of cash.
Stanford geneticist David Botstein, a basso-voiced firebrand, stormed the podium, saying that the whole idea was a senseless waste of resources. Only so many science dollars were out there, and that sucking sound you would hear from sequencing the human genome signified far worthier projects disappearing down a black hole—indenturing scientists, especially young ones, to some monster project that would rob them of their own creative efforts. Loud applause.
Thus, the project looked doomed. Except it wasn’t—not entirely. For the next two years, the idea managed to crawl along with just enough support to keep it from entirely vanishing. And then one day James Watson agreed to run the project. Instantly, everyone snapped to attention.
“A mover of people” was how Watson saw himself. And that was a fair description, because when it came to the nexus of genetic research and academic politics, he was, indisputably, the 800-pound gorilla. Thus, on October 1, 1990, the announcement was made in cooperation with researchers from Great Britain, Japan, Germany, China, and France. With a billion dollars in federal funding and Watson at the helm, the HGP was officially launched. A 15-year deadline was set to complete the project, now titled the National Human Genome Research Institute (NHGRI). At least that was the plan. Plans, however, sometimes have a way of going sideways.
* * *
—
AT FIRST THE WORK BEING DONE at NHGRI was fine and good from Craig Venter’s point of view. But as time passed, the sluggish pace at which the project was moving increasingly got under his skin. And after a while, he began to make his opinions known.
In the mid-1980s, Craig Venter was just another among the battalions of researchers at NIH. He was considered a good enough biochemist, but hardly a man who could strap it on with the likes of James Watson. But now, 10 years later, while Art Levinson was busily moving up Genentech’s corporate ranks, and Ray Kurzweil was making a national name for himself as one of the country’s most promising inventors, Venter’s career had advanced too.
The big leap came when he ran across his first automated DNA sequencer: a machine that could identify genes by shining lasers on all four DNA bases, instantly revealing whether they were an A, C, T, or G, and then record the information sequentially in a computer. Right away, Venter saw that the sequence of each gene could be recorded far more rapidly with this machine than any other technology. Immediately, he boarded a jet and headed to Foster City, California, to meet with one of the instrument’s creators, a scientist named Mike Hunkapiller. Such a machine was a beautiful thing to behold. Venter didn’t buy one. He immediately bought two.
Sequencing the human genome the way Hunkapiller’s machine did, inside a computer, meant scientists could now theoretically track all the genome’s linked letters in the order they existed—something like an extraordinarily long list of telephone numbers—and they could do it at blinding speed. And yet, even after Venter shared this new discovery, the approach wasn’t adopted at NHGRI.
The reasons were complicated.
By 1993, a new scientist had taken over for James Watson at NHGRI. His name was Francis Collins. Collins was considered an all-star gene hunter, and a man of serious scientific distinction. Even though he was only 43 years old, he ran a richly funded genetics laboratory at the University of Michigan. In collaboration with other scientists, he had used a method called “chromosome jumping” to uncover the genetics behind cystic fibrosis, Huntington’s disease, and Hutchinson-Gilford, an affliction that rapidly aged its victims.
Right out of the gate, Collins had his hands full at NHGRI. He could already see that the HGP was never going to hit its self-imposed 15-year deadline. And despite Venter and Hunkapiller’s new, faster approach, the institute still wasn’t certain that was the best way to accurately sequence the genome. Collins felt Hunkapiller’s machines were too sloppy, and he believed, adamantly, that the project’s accuracy should never be sacrificed for the sake of speed. A standard was set that had to “stand the test of time,” he said: No more than one base pair could be misspelled out of every 10,000. And that was that!
This drove Venter crazy. The central vector in his universe—the one that struck at the very heart of his view of science, and really his view of life—had been drilled into him by his mentor Nate Kaplan at University of California, San Diego.
Every scientist, Kaplan told him, had to “do the experiment.” They had to be willing to try new solutions to old problems. Yet so many were afraid to try anything new. Why? Because in their heart of hearts they feared what they might find: the horrifying truth that their theories, their ideas, the comfortable concepts that they had wrapped their minds around for so long, might be wrong! What if they tried something new and it shattered their hypotheses? So rather than move forward, rather than push the envelope, too many scientists found it convenient to avoid the experiment and cling to the status quo.
Venter never forgot Kaplan’s words. He made The Experiment a fire-breathing metaphor for his life. Take chances. Knowledge was power, even if it obliterated what you held dear. It was the only way science, humanity, or anyone at all could hope to advance. And it was the best way to meet the holy promise he had made that day on China Beach: Make something worthy of your life!
Nevertheless, Collins held to his “must stand the test of time” standard. It was slower, but at least it was reliable. It wasn’t that either scientist cared about the undertaking more than the other. Both viewed it as one of the great—perhaps the greatest—scientific endeavors ever. Collins himself considered it bigger than splitting the atom, and he had dubbed the human genome the “book of life” in his lectures. Yes, they could both agree the project represented noble, high-minded, historic work. But that was where the common ground ended.
Whole books and uncounted articles have been written about the political and scientific knife fights that ensued during this period. But in the end, it came down to this: With backing from PerkinElmer, a big, Connecticut-based analytics and computer science company that had brought in Hunkapiller’s gene sequencing company, Venter created Celera Corporation, a for-profit company that could, all on its own, sequence the human genome. T
he company had all the backing it needed to complete the entire job: $300 million. In short, Venter was ready to move ahead, and he didn’t need NHGRI or Francis Collins or anyone else to make it happen.
In a dramatic meeting at the Red Carpet Club at Dulles Airport on the afternoon of May 8, 1998, the two scientists faced off: Venter, the California kid and ex-Navy corpsman with the blue-volt eyes, now bald and shorn of his golden California locks but still carrying the outlines of his swimmer’s body, and Collins, lanky and bookish, with a great mop of dark brown hair sweeping across his large forehead.
Venter was respectful, but direct. “We don’t think people want to wait another seven years for you to finish the genome,” he said. “Perkin Elmer and I have teamed up to form a new company. Our goal is to do it ourselves, using a couple hundred of Mike’s new sequencing machines. We are going to make the genome free and available to everyone, same as you. The main difference is, we estimate we’ll be done in 2001, four years ahead of your schedule.”
Venter’s scheme was a huge gamble. Nothing like this had ever been attempted before—certainly not in the annals of biology. Every technical component had to work seamlessly, or the whole undertaking would go down in a howling ball of public flame, with Venter hung out to dry like a sack of wet laundry. But that was okay with him. Maybe the entire endeavor would fall flat on its ass, he told Collins. Maybe not. The point was: It had to be tried! Do the experiment! And that was precisely what he was going to do.
* * *
—
CELERA DIDN’T COMPLETE sequencing the human genome in three years, by 2001; it did it in two—by July 1, 2000, to be exact. During that time, all parties—NHGRI and Celera—eventually negotiated an agreement, and announced that the genome had been sequenced as a joint effort. Everyone bought in: Venter, Collins, President Clinton, and all the other organizations involved from around the world. The big day, at last, had come.
Six hundred members of the world press arrived on cue to get the word: TV crews with their glaring lights and lenses; journalists from the most influential magazines and newspapers, notebooks and recorders poised, eyes wide; photographers and assistants, all scrambling like some immense, many-legged beast to learn how science had accomplished what the president of the United States himself described as “the most important, most wondrous map ever produced by humankind”: the map of the human genome.
Venter had waited a very long time for this day—fought for years, viciously, some said, although that was a matter of opinion. One man’s vicious was another man’s tenacious. He had been vilified, marginalized, and labeled crazy—but in the end, he had found a way. More than one among the gleaming towers of academe had called him a cantankerous megalomaniac hell-bent on destroying the high-minded aspirations of the original government team while seeking to place himself in the limelight. One scientist, in a New Yorker article, had flat out called him an asshole! But it was either create Celera or sit on the sidelines like some spear-carrier as the NHGRI dawdled and threw valuable time and money away.
Still, Venter would have preferred to be well liked and appreciated. He had said again and again that he didn’t want the approval of NHGRI’s top brass, but that wasn’t entirely true. One of the great ironies of Venter’s life was that he craved the approval of the very people whose noses he so often thumbed: his older brother, his father, the academic insiders, and the anointed.
But today, at least for a little bit, he had it all. He had stood at the right hand of the president himself. He had done the experiment, and on the big day said his piece, right there in the East Room where Meriwether Lewis had once sat with Thomas Jefferson to map another vast and mysterious place: the Louisiana Territory. Now, he was certain that by sequencing the genetic code—this remarkable string of molecules that made the human race possible—science could at last get down to the business of taking on the diseases that unraveled us all: cancer, diabetes, Parkinson’s, Alzheimer’s, the whole messy lot. Countless lives would be saved, and that, he was certain, would absolutely transform the medical world. It was just a matter of time.
12 | THE ACCELERATION OF ACCELERATION
March was one of the nicer times of year to be in Washington, D.C.—as long as you discounted the politics. It was generally free of the oppressive swamp heat and shirt-soaking humidity of summer, when the throngs clogged the National Mall and the endless procession of tour buses belched and threaded their way through the capital. In early spring the weather grew mild with the cherry blossoms on the cusp. But that wasn’t the case on March 14, 2000, as Ray Kurzweil blithely made his way to the East Room of the White House wearing a dark suit and a muted red tie. Today it was cool, with a stiff breeze out of the south blowing bright cumulus clouds above the city’s grand monuments.
Life was good for Raymond Kurzweil as he dipped his toes into the shallow waters of the 21st century. He was on his way to spend some time with the president of the United States, William Jefferson Clinton, who would that day hang the National Medal of Technology around his neck. Not only was he being feted at the White House, but he also had finally begun fusing his prescriptions for immortality and artificial intelligence with the publication of his latest book the previous October, The Age of Spiritual Machines. That was 29 years after his father’s death and 32 years after he created his first computer. Apparently, even for Ray Kurzweil, big ideas took time to percolate.
Kurzweil wouldn’t be alone that day when he received his medal, of course. He would be joining 16 other scientists who were also receiving theirs. (One award was posthumous: Bob Swanson, co-founder of Genentech, who had died of brain cancer only a few months earlier.) Craig Venter would not be among the awardees either. He wouldn’t win his National Medal for another eight years. Nevertheless, all of the explorers assembled in the room that day could feel Venter’s presence. Three and a half months from now he would, at the very same podium, in the very same room at the White House with Clinton, make his human genome speech—the one the whole world would watch. Anyone listening closely enough, in fact, might have thought the National Medal ceremony was more about the Human Genome Project than the collective accomplishments of the scientists assembled. On that day, Clinton—before distributing the heavily ribboned medals—announced that the United States and Britain had agreed that very morning that all genetic information, once the Human Genome Project finally shared it, must be, “Free to scientists everywhere for people everywhere.” Clinton felt this was crucial because the advances would not simply change science; they would change everything.
The project’s findings were already revealing the genetic links between scourges like leukemia, schizophrenia, and kidney disease, and Clinton felt that soon they would do the same for cancer and heart disease and all the rest. The mysteries of human weakness and mortality, he said, would be laid bare, and then the solutions would follow.
If it had crossed their minds, any number of the bemedaled attendees in the East Room could have switched out the names of all the maladies Clinton had just mentioned and substituted only one: aging. Had they done that, they would have glimpsed an insight that had already begun simmering in Ray Kurzweil’s mind five years earlier, in 1995.
The insight was simply this: Unless a meteor struck Earth, the human genome was, one way or another, going to be sequenced. What’s more, it was obvious that the genetic information pouring out of the HGP was doubling every year, while at the same time the cost of its creation was dropping by half. This meant advances in computer science and biology were merging; in their way, humans and machines had begun to meld. He knew all of this because he had written about it in his books.
So now, as he sat perfectly erect with the other honorees in front of the podium, smiling up at the commander in chief that March morning, Kurzweil thought to himself, They are finally getting it. And of course they had to because the Law of Accelerating Returns made it so.
* * *
—
THE LAW OF ACCELERATING RETURNS, or wha
t Ray Kurzweil came to call LOAR, had begun to dawn on him as early as the 1980s, but wasn’t really codified until the publication of The Age of Spiritual Machines. It started out mostly as a practical matter; he needed a system for timing the rollout of all of those technologies he was continually creating. In Kurzweil’s universe there was never a shortage of concepts—but what good did it do to conjure a good idea, or even a well-executed invention, if there wasn’t a market for it? Timing was paramount.
By the 1980s, Kurzweil had gotten pretty good at foreseeing the future. He predicted a computer would beat the world’s best human chess players by the year 2000—and lo and behold, in May 1997 it happened. IBM’s Deep Blue computer defeated Garry Kasparov, the chess world champion, in one of the highest profile competitions ever. The world gasped.
Kurzweil also foresaw the explosive growth of the internet in the early 1990s, when the world’s total population of users was a mere 2.6 million. In 2017, that number would clock in at 3.7 billion, more than a thousandfold increase. Smartphones, cloud computing, and self-driving cars were also among his predictions. Not that he was always right, but he clearly foresaw something in this idea of digitizing the human genome, and all the exponential business that went with it.
In his efforts to predict the future, Kurzweil had turned to Moore’s law. Gordon Moore was one of the founders of NM Electronics, which later became the Silicon Valley juggernaut Intel Corporation. By 2000, it easily stood as the world’s most dominant manufacturer of advanced silicon chips. In a 1965 article for Electronics magazine, Moore noted, almost in passing, that the number of components in the integrated circuits of the day—things like transistors, diodes, and capacitors—appeared to be doubling every year, and probably would continue to do so for at least the next decade. (In 1975, he amended his insight, and changed the rate to every two years.) The point was: Change—at least when it came to integrated chips—was advancing, exponentially. This discovery became the very foundation of Silicon Valley and the explosion of ideas, money, and transformation it generated over the next 40 years.
