Propelling both business and science, Boger had always behaved with the easy, sweatless confidence of a grand master playing chess with a dozen or so opponents in a show match. Now, with Vertex pursuing Schreiber, an IPO, a drug, a new project, and publicity for its first important publication, he seemed challenged for the first time. It was as if the games suddenly interlocked in three dimensions. A move in one varied the play in all. The time for each move was halved. Real money was on the table.

On Tuesday, May 7, 1991, less than a week after deciding to take Vertex public, Boger arrived in Cambridge before 7 A.M. with two goals utmost in mind. The previous Saturday, Schreiber had mentioned in a talk at Yale that his FKBP-12 structures would be published in Science that coming Friday. For Vertex to be able to claim a tie on the X-ray structure, that gave Yamashita and Navia a deadline of no later than noon Thursday, when advance copies of Science would be released to the media. After that, their assertion of independent discovery would be irrevocably stained. At the same time, Boger fretted being shut out in the initial burst of publicity that no doubt would follow Schreiber’s papers. Moore’s structure was scheduled to appear in Nature the following week, but Nature had a prepublication embargo that prohibited researchers from announcing their work without its consent. Violating the embargo and jeopardizing Vertex’s relationship with the journal were out of the question. Also, now that Vertex was going public, it was barred from seeking any publicity that might inflate its stock price. Not yet technically in the federally mandated “silence period,” the company nonetheless had to be careful about what it disclosed lest it delay Kidder’s tentative offering date of mid-July until August, when Wall Street goes on vacation. Boger knew that sooner or later, probably sooner, the market in new biotech offerings was going to crash. He shuddered to think about waiting until September to go out.

He spent most of the day cloistered with Aldrich, Ken, and the Kidder people, working on the red herring, the prospectus that Vertex would submit to the SEC and eventually investors. At 10 P.M., he was still at his desk. With less than forty hours left, Navia, Murcko, and Yamashita were racing furiously next door to reconstruct the protein.

“We’re going to have to work like crazy to get FedExes out tomorrow and get Nature to break its embargo and get the lawyers to give us an opinion that we’re not illegally hyping the company so that the SEC won’t put back our offering,” Boger groused convivially. “It’s ugly, I hate it, but it’s part of the business.”

Murcko, overhearing his lament, muttered, “We’re going to have to get that guy back to being a scientist.” But of course the matter was moot and, in Boger’s mind, beyond distinction.

•  •  •

“Are you going to fire me for this?” Yamashita asked Boger at around 8 the next morning. He had been working all night.

“I have to wait for the data,” Boger chortled.

Typically, Boger was 70 percent joking and half serious. He in fact blamed Yamashita for very little, thought he had done an extraordinary job under crushing circumstances, and would support him even if he failed again to get the correct structure. On the other hand, having X-ray structure mattered immensely now and Boger wasn’t paying Yamashita, who had bounced back from his moral crisis with characteristic resilience, to fail indefinitely.

Far more urgent was the exquisite irony of Vertex’s public relations conundrum. For almost two and a half years, Boger had sold the company’s story mightily, doing all that he could to make it sound solidly grounded in science. He’d “worn out three pairs of tap shoes,” he said. Yet now, at precisely the moment when it had a major discovery to announce; when Business Week and Fortune, in their expanding coverage of the biotech boom, had discovered Boger and were planning articles about him and Vertex; when the publicity benefits of positive media coverage would be incalcuable as the company went public, he was forced to be guarded, silent. To the salesman and the scientist in him, the incongruity was painful—“smothering the baby,” he called it.

“I blew off Gene Bylinsky from Fortune this morning,” he told Aldrich, gulping exaggeratedly in disbelief.

“Goes against the grain, doesn’t it?” Aldrich said.

“It’s galling. Anybody who wants to get around the SEC, to sell junk stock, to bilk widows and orphans out of their life savings, is not going to be hurt by these rules. The only people who are hurt by this are the ones who try to play by the rules. It’s not protecting anyone. It’s enough to turn you into a raving conservative.”

Boger’s first priority was to get Nature to lift its embargo. He might not be able to tell Fortune that Vertex was hard after Merck in developing a drug, but there was nothing in the SEC regulations to prevent the company from issuing a press release about a vital discovery with clear importance to the public health. The issue there was simply, if critically, one of timing. If Vertex announced Moore’s structure at the same time as Schreiber, Karplus, and Clardy announced theirs, the media would report a tie and Vertex could exploit it accordingly. If Vertex was forced, however, to withhold its announcement for even a week, there would be no coverage to exploit, just the crumbs of a few second-week stories in the science press. Moore began the morning by calling Nature’s Washington office. Reminded that the journal’s embargo was meant to discourage exactly the kind of commercial exploitation Vertex was proposing—exploitation that Nature believed did nothing to advance scientific inquiry—he charged back. He explained that Science was about to publish back-to-back structures of the protein. An hour later, a Nature editor in Washington called back to say the journal had changed its mind.

Moore sped to tell Boger, who called Ken, who told him that he and Kidder’s lawyers, who were paid to be skittish about such things, now agreed that the SEC had no case to delay Vertex’s offering. Hanging up, Boger strode triumphantly to the lobby, where a stack of press releases was waiting to be faxed to the Wall Street Journal, the New York Times, the Los Angeles Times, the Washington Post, the Boston Globe, Business Week, and perhaps two dozen other trade, industry, and scientific journals. “The blockade is down,” he announced. “The ships are coming through. I want to blitzkrieg right now. When the press accounts come out, I want the footnote to say, ‘There are also some people at Harvard who claim to have done this.’ ”

Boger hadn’t won, but he knew now he hadn’t lost. Reflecting on catching Schreiber, his overall glow was only slightly less boistrous than if he’d beaten him outright. “It’s not quite what I want,” he explained, returning to the lunchroom. “I want to rub his nose in the dirt and step on his head. But I’ll settle.”

To the scientists, Nature’s second reversal in a month was illuminating, improbable, and not without its jealousies. “This is the value of working in hot areas,” Moore told Harding. “Science and Nature have no idea whether any of these structures are right, but they’d obviously rather publish shitty structures of important proteins than exquisite structures of boring proteins.” Swept up in the tide of anti-Schreiberism, he added, “We’re going to spoil his party. We’re going to be a very sharp dart in his ass.”

Harding was sullen, dismal. While the spotlight had shifted to Moore and Yamashita, he had been working since January to identify other FKBPs. Characterizing one at last, he’d just learned that morning that SAB member Steve Burakoff had a paper in press identifying the same protein. “They want to do their work, that’s fine,” Harding grumbled, referring to Burakoff and Barbara Bierer, Burakoff’s chief collaborator and a Vertex consultant, “but there’s never any sharing of any information.” He was now working furiously to file a patent application to establish a priority date before Burakoff’s paper appeared in print. It was another instance of Vertex being beaten by one of its own advisors, and Aldrich, furious, now wanted to fire the entire SAB.

Harding, long-suffering and perplexed, shuffled back to his desk, a pale shadow of Moore’s and Boger’s fire-breathing personae. It was to be for him another long, lightless day.

•  •  •

Throughout the morning and into the afternoon, Mark Murcko, Vertex’s chief molecular modeler, performed “dynamics runs” on the protein. Now that he had a refined crystal structure of FKBP-12, the key questions were, What does it do and how does it do it? A simple snapshot wouldn’t do. Murcko had to simulate the molecule’s natural activity, bring it alive, posit its existence in not only space, but time. He had to “feel” it. “Karplus’s group did this a month ago,” he said. “They had twenty postdocs and access to an order of magnitude [ten times] more computing power. It’s that against basically a couple of us, but who am I to be different?”

Sitting at a workstation diagonal from where Yamashita was completing his refinements, Murcko programmed it to simulate the protein’s natural wriggling. Atoms gyrate in nanoseconds—billionths of a second. To slow them down, Murcko used the computer like an ultrafast strobe, freeze-framing every 0.3 picosecond (0.3 trillionth of a second). He then ran the images back in slow motion. They resembled the “washboarding” of bees, a herring-boning, a shimmy. The entire dance lasted just a few seconds, though in it Murcko thought he saw hints of how the molecule might work.

He was especially interested in the active site. Most of the molecule shuddered only slightly; it was highly constrained. But at the opening of the hollowed-out core that Navia and Yamashita speculated was its “business end,” the place where its atoms interacted with other molecules, including FK-506, Murcko noticed a remarkable flexibility. On one side was a group of atoms that he compared to a flap. At one point during the run, about midway, the flap swung open perhaps thirty degrees, like a miniature gate, then snapped shut. The implications for drug design were profound. If true—and Murcko was quick to volunteer his usual reservations and disclaimers—this movement in the flap area suggested several possibilities about the enzyme: that in binding it could accommodate larger groups of atoms than the size of its opening would normally allow; that the flap might act like a trap, holding the bound molecule in place once it had infiltrated the active site; most tantalizing, that the shape of both the protein and the small molecular binder, such as FK-506, changed significantly as the flap curled around the inhibitor. Murcko began to speculate that unlike in Schreiber’s hypothesis, which was that part of the drug bound to the enzyme while the rest reached out and interacted with another protein, there was a subtler geometry at work. He thought the combined entity of drug and protein might form a new shape and that the new conformation accounted for its action.

In the race with Schreiber, Murcko’s simulations provided a potent new thrust. Vertex had hoped only to catch Schreiber, but now, as Navia put it, “We’ve got our own story.” Navia, particularly, was elated. “This paper is going to be a killer,” he told Murcko. “It’s going to be a lot more interesting than I had dared believe.” Privately, he began to look forward to a full-scale debate with Schreiber, a high-profile slugfest that would avenge the split between them and climax, most probably, in a head-to-head confrontation at the FK-506 conference in Pittsburgh. There, with the scientific world watching, Vertex for the first time might challenge Schreiber’s priority in the field by offering a new theory on the central issue of how the protein worked. At the very least, it now had something to offer as scientific interest shifted to structural biology.

The deadline for submitting the paper—it was now Wednesday afternoon—was less than twenty-four hours away. Yamashita was scrambling to finish the structure, though in fact his contribution—and the pressure he had felt—had already begun to recede. All he had left were a few refinements around the active site and to account for the water molecules in his density. Otherwise the onus was on Navia, who had worked until 3 A.M. the past two nights drafting the paper for Nature and would shepherd it through to completion. It was strange. Yamashita had wanted to solve the structure of the protein more than anything in the world. He had pushed himself dangerously for more than a year to do it, been beaten, recoiled bitterly, recovered, blundered badly, collapsed, recovered again, and was now within a day of being lead author on what was likely to be a career-making paper—a paper that Navia, his “nemesis,” was writing for him and that seemed all but certain, based on prior agreement, to be accepted into one of the world’s premier scientific journals. He had gotten, it appeared, almost everything he wanted, yet the pain of getting it had been so great that he felt no pleasure, only a comforting distance, as though he were alone in a life raft drifting from the scene of an excruciating disaster—the wreck of his own life. His prayers nearly answered, he was wary, numb, cynical, and oddly composed.

Besides cleaning up his final structure, he had one major matter left—authorship. As lead author, it remained Yamashita’s responsibility to name those who would share credit with him, a choice that had previously opened rifts that he now hoped to close. He and Thomson had shakily repaired their friendship after their fight a month earlier, but Yamashita had not brought up the issue, and Thomson had continued to insist that he didn’t want his name on the paper. Meanwhile, Thomson and Navia had reached an icy standoff, with Thomson complaining about what he considered Navia’s ineptitude with the protein and Navia insisting that Thomson was blackmailing crystallography by holding protein hostage. Working again around the clock, Thomson had developed a severe case of conjunctivitis and wouldn’t be in until after four—an hour away.

Yamashita sought Boger’s counsel. He wanted Moore and especially Thomson back on the paper, though he understood Boger’s determination to get the scientists to think less about individual credit and more about drugmaking. Boger was snappish; he’d heard Navia’s complaints, and though he didn’t believe them, he wanted to make sure Yamashita wasn’t including Thomson simply to pacify him. “I own you and I own John Thomson, and there’ll be no quid quo pros here,” he said. Yamashita responded that he also didn’t think Thomson was witholding protein, but that he wanted to share authorship with him so that Thomson would continue to talk and work with him. “If you do that,” Boger said, “John’s name won’t be on the paper and John will be fired. Tell me how that’s different from going into your dry cleaners and saying, ‘This is a dangerous neighborhood, and it’s going to be even more dangerous if you don’t give me $500.’ It’s extortion.” Yamashita left more confused than ever.

In the modeling room Navia was asleep in his chair, sitting upright, his tie still straight, looking comfortably ursine with his broad chin sunk to his chest. He was snoring lightly. Nancy Stuart came in. Stuart, Aldrich’s chief assistant, had become Vertex’s all-purpose factotum. Monitoring the project councils, developing marketing strategies, she strove to keep the scientists on task. At thirty-two, she was the person on the business side the scientists most trusted, having worked for a time at the bench herself and having a natural empathy for people. Gently, she tried to shake Navia awake. He didn’t budge. She tried again, Rolfing him this time.

“I need to talk to you about our Wall Street Journal strategy,” she said. “Everything has to go through [enterprise reporter David] Stipp. But you’ve got to call Waldholz.” Veteran medical writer Michael Waldholz had covered the HIV protease story for the Journal and was one of the few reporters any of the scientists, including Boger, knew personally, much less respected. Sluggishly, Navia removed himself from his seat, from science, and placed the call.

Yamashita stared at a new field of density. He could identify the active site in his sleep, but he was looking for something else, evidence of Vertex’s lead molecule, 367. By itself, the structure of FKBP-12, much as it had nearly destroyed him, was insufficient for drug design or figuring out how the molecule worked. The ultimate test was to solve the structure of the enzyme bound by other molecules, so-called complexes—in other words, to show the lock together with a variety of keys so that researchers could see and compare exactly which points made contact. Schreiber and Clardy had apparently already solved the complex of FKBP-12 and FK-506. Having complexes with Vertex’s own compounds would show what changes the chemists might make to improve the activity of their molecules.

There were two ways to grow complexes, equally devilish and uncertain: soak the drug into existing crystals or, more tryingly, coax them into crystallizing as a unit. Yamashita had tried first to soak in the drug, but now, as he looked at the results on the screen, there was no additional pocket of density in the active site, no unaccounted-for netting. The soaking experiment had failed. Discouraged, Yamashita reported the result to Murcko, then left immediately for the cold room to begin trying to crystallize FK-506 and FKBP-12 together. “We know that it must work because Stuart’s done it,” he shrugged. “Anyway, its the logical next step.”

“It’s a serious blow,” Murcko said, looking up painfully from his simulation. “Instead of going home tonight at midnight and having the first glimmerings of where 367 sits in the active site, we don’t know. I have no idea when I’ll know. Is it a week? A month? Three months? Six months?” He stopped himself, frustrated. After more than a year at Vertex, he still didn’t have the information he needed to design with any real conviction new molecules, and the gap between Vertex’s promise and its reality had begun to cause him to complain more than he liked. Science was tenuous enough without one’s having to doubt whether one was a fraud. “I’ll continue to do things to be useful,” he murmured, “but in the absence of real data, anything I’m going to say is going to be speculative.”

•  •  •

Boger was in command mode, managing a half dozen issues at once, another half dozen roiling at bay. He was jocular, intense, flippant, stern, unfiltered. He was moving from chessboard to chessboard with rifling speed. At his desk, the fingers of his right hand rolled over his turbomouse like a magician’s over an eight ball as he stared imperturbably at the big screen of his Macintosh, his looking glass, his muse. He juggled phone calls—with Ken, with Holman, with Schmidt, with Kinsella—with round-robin visits from the scientists, Aldrich, and Nancy Stuart.

His ego, never restrained, was incendiary. To Ken he suggested calling the Harvard patent office “to put the fear of God in them.” Vertex no longer had any claim on Schreiber, but now that it had pulled nearly even with him, Boger wanted to spray his path with tacks by suggesting that the timing of his agreement with Clardy might have violated his Vertex contract. Flipping through another company’s red herring, he dug into an imaginary trench coat. “ ‘Pssst. Hot area. Inflammation,’ ” he sneered, then spat, “It’s a technological Ponzi scheme written by investment sharks . . . It’s Florida land that’s land only six hours a day.” Talking to Aldrich, he resolved to tell Burakoff: “We consider what he’s doing a serious breach of contract and he’s worthless to us as an SAB member as long as he’s working in the same area.” Taking pleasure in warning Kinsella that he might have to cancel a flattering Wall Street Journal article about him—a profile that Kinsella had eagerly coveted to promote his new company with Schreiber—he cackled when Schmidt, on another line, couldn’t remember Kinsella’s name, referring to him as “that guy on the board who put in the least money.”

Knowles, who had seen Boger this way ever since he was in graduate school, demurred in his call to Boger, “Perhaps I’d better be out of town for a few weeks.”

Boger popped into the lunchroom. Ever since he’d started Vertex, he’d intended the area to be like his kitchen when he was growing up, a hub. It was here that the scientists, each reeling in his own orbit, took their pit stops. Scarfing some Ritz crackers and a juice, standing as always, Boger sidled toward Yamashita, who sat picking through the Times. Towering benignly, he told him the price of the chair he’d smashed was $280 and that it would be deducted from his paycheck.

“When I was a kid, each of my brothers and I had a spitting tree,” Boger said. “When we were really mad, we could spit. I used to slime that tree all the time.”

“I like to physically destroy metal,” Yamashita said.

“You’re too high-tech.”

“Maybe. My lab chief in graduate school gave me an old terminal and a sledge hammer to smash it with,” Yamashita said. “It was wonderful.”

“You look relaxed now.”

“And I didn’t even have to destroy anything.”

It was a loose, giddy moment, the first between them in weeks, and Yamashita seemed relieved to be able to joke about the violence that had only recently threatened to consume him. Indeed, nothing now was taboo. When Navia came over to congratulate him on his refinement, saying “These geometries are fabulous,” Yamshita replied, “Or a fabrication.”

With Boger providing the emotional cues, there followed an outbreak of verbal riffing that culminated, inevitably, with a mass skewering of Schreiber. The structure of FKBP-12, though it had been solved by Schreiber’s and Karplus’s graduate students and by Clardy’s group at Cornell, was clearly going to be Schreiber’s moment. By the rights and rules of science—credit accrues upward to the senior investigator; initiating an idea counts more than proving it—even the X-ray structure, which Schreiber had contributed to mostly by providing protein and synthesizing heavy-atom derivatives, was likely to be known as “Schreiber’s structure.” His string of path-breaking discoveries with FK-506 remained unbroken, and his crossover into biology was progressing even better than he’d hoped. And yet on the day of perhaps his most notable achievement, he was about to find himself sharing honors with an upstart company that had ingloriously fired him eight months earlier, a place where he was reviled. The scientists loved it, though clearly not as much as Boger and Aldrich, who gloated radiantly.

“Imagine Stu-Bob [Boger’s and Aldrich’s name for Schreiber] driving down Storrow Drive in his Porsche on his way to Harvard, hearing it on National Public Radio,” Aldrich said. “He’ll wrap it around a tree.”

Navia picked up the theme. Humming the familiar introduction to “All Things Considered,” he pretended to be driving self-satisfiedly. Suddenly his face contorted: “Screeech! Craaash!”

Only Thomson refused to join in. Sitting away from the group, glowering at Navia’s clowning, Thomson nursed his conjunctivitis and a painful upper body rash that afflicted him during long sieges of cutting tissue and washing glassware. “Nobody appreciates what we have to go through,” he said. “A lot of unpleasant, up-to-our-armpits-in-slop work that’s caused a lot of ailments [he pronounced it “I’ll-ments”] that if we were in an industrial setting would qualify us for workman’s comp. For every one hundred milligrams of protein we produce, we have to process a fifty-five-gallon drum’s worth of phosphylated thymus juice with a lot of nasties in it in one-pint chloroform jars. But we do it.

“I’d like to know,” he muttered, “what some of the others do.”

•  •  •

“Josh and I don’t want to be quoted,” Aldrich told the disembodied voice of Vertex’s PR man on the speaker phone in his office. “Nothing about billion-dollar drugs. When calls come in, if they ask for me, the front desk is going to say I’m unavailable. If they ask for a scientist,” he nodded to Moore, who had been corralled for the briefing, “it’ll kick over to you.”

Aldrich wasn’t opposed to exaggerating claims on principle, but he knew that Vertex had to be exceedingly careful about how it handled press inquiries about Moore’s paper. Moore sensed his caution. Though he thought the rehearsal unnecessary and disdained being coached by a PR person whom he couldn’t see, he was eager to cooperate. Like a rookie pitcher in his first outing, he thought he’d play it safe. “I’ll just say it’s an important step,” he volunteered.

“Facts, not suppositions,” Aldrich advised. “You can say something like, ‘This is an important step, but it’ll be several years before there are drugs to show from it.’ ”

The PR man counseled, “You don’t want to be in the position, Jon, of saying ‘No, no, no. I can’t answer that.’ ”

“I could just tell them research itself is never patentable.”

“You don’t want to say that,” Aldrich interrupted, joking. “You never know what we may want to patent around here.”

Aldrich and Nancy Stuart had prepared a backgrounder, explaining the importance of Moore’s discovery and answering hypothetical questions. The PR man complimented them and, as was his job, intimated great things.

“We’re going to put it on the wire so everyone and his grandmother will know about this,” he said.

“That’s OK with us,” Nancy Stuart said, smiling. Aldrich rubbed his chin thoughtfully. Moore tried not to roll his eyes.

•  •  •

Sitting next to Navia in the X-ray lab at around 7:30 P.M., Yamashita hastily concluded that the solution to the authorship problem was to remove his own name from the paper. Quickly, he reconsidered. “Camus says it’s stupid to try to make a point to the world by committing suicide,” he said. “In a likewise fashion, if they give me first authorship and even if I might feel very strongly I shouldn’t take it because other people are being screwed, does it help for me to say no?

“I may be unexpectedly ill tomorrow.”

Navia leaned over from his microscope. “One of these moments,” he said, “when you get a chance, I’d like you to see what a spectacular job you did.”

He was scanning Yamashita’s cocrystallization attempts with 367, not those from that afternoon, but others from earlier in the week. In several of the wells, he saw incipient needlelike crystals growing in clear formation. Unlike a year ago, when Yamashita’s first putative crystals of FKBP turned out to be salts, Navia had no doubt these were protein.

“Oh, man. That’s crystalline.”

“Do you think so?” Yamashita said, tempering himself. “We’ll see tomorrow.”

“No, Mason, I think you’ve got it.”

Slapping Yamashita on the back, Navia left to rejoin the paper-writing effort. Obliterating all further thoughts of martyrdom, Yamashita shrugged.

“That’s the way life goes,” he sighed. “I’ve learned from being here, you just find the next experiment to do. That’s what keeps us from destroying ourselves.”

•  •  •

Boger charged from his office a half hour later with a first draft of the Nature abstract. A summation of no more than one hundred words, it was the most critical part of the paper, the “take-home message” most readers would scan as they foraged through the journal. Boger, who’d taken several stabs at it while Navia updated the text, handed it to Navia without a word.

“How far do we want to put out our necks?” Navia asked.

“What ‘we,’ Kemo Sabe?” Boger said. It was yet to be decided whether Boger’s name would appear on the paper, though as a general rule, he, like Tishler, opposed being named except in those cases where he’d made a direct contribution. He was less interested in his résumé than in the larger prize of producing a drug. Credit, he thought, was the prerogative of those who did the hard intellectual work of eliciting data, not those who simply set it in motion or interpreted it comfortably afterward. A chemist, he applied the chemist’s restrictive code for deciding whose name to put on a patent; either you drew the cartoon or you made the molecule. Everything else was superfluous.

It was after 9:30 when he, Murcko, and Navia stopped to eat. Tearing into containers of pan-fried dumplings, pan-fried vegetables, and egg rolls, they ripped with equal ferocity into the draft. It was the first Boger had eaten all day.

“Here’s what I’m after,” he said, legs swirled, pinching a dumpling with his chopsticks, stabbing the air for emphasis. “Stuart already owns the binding domain, effector region hypothesis, but I don’t hear any enthusiasm around here for that hypothesis. I think the answer is going to be that something about it is right, but that it’s not enough. I want us to stake out the alternative hypothesis. I want us to put forward the last great hope of biological explanations, a hypothesis through which you may be able to drive a truck-train later on but that stakes out new ground now.”

Navia told him about Murcko’s latest simulations in which he’d modeled one of Vertex’s best inhibitors into the enzyme’s active site. “It may be too small,” he said. “The flap may be coming in and crushing it. We need a doorstop for the flap.”

“A molecular doorstop,” Boger said tantalizingly, as if finding the light switch in a dark room. “It’s a larger hypothesis than Stuart’s, but entirely consistent with it. Not only do you need an extended effector region, but you’ve got to hold the flap open just so.” In other words, Boger speculated, it was not just the atoms that stuck from the bound conformation like a gaff that accounted for FK-506’s biological activity, but the way that those atoms changed the outer shape of the protein, like a tongue pressing hard inside a cheek. It could be a critical observation for drug design as well as a direct challenge to Schreiber, and Boger, pleased, was comfortable with asserting it, even though Vertex hadn’t proven it.

“The sin in science is not making mistakes,” he said, hungrily spooning out more vegetables for himself. “It’s overinterpreting your data. If you’re disproved by a massive amount of new data, that’s fine. What you don’t want to be is disproven by a little more data. That means you went almost but not far enough.”

Abandoning the mess in the lunchroom, Boger, Navia, and Murcko returned to their computers to justify their latest observation. With fourteen hours to go, they would do the next set of experiments.

“This is really enjoyable,” Murcko said, knowing he would now be working through the night and right up to the next day’s deadline, scrambling to produce simulated data to support a speculative theory that had been hatched spontaneously just moments before, “in a perverse sort of way.”

•  •  •

In the chemistry lab, Roger Tung and Dave Deininger labored against less theoretical constraints. For weeks they had been working most nights until midnight, trying to make new molecules to inhibit HIV protease. Unable to sell the AIDS project in Japan, Boger had also been unable to expand it, and Tung and Deininger had chiefly borne the consequences. The company had now poured $2 million out of capital into HIV research, and the project was foundering in a kind of catch-22: They couldn’t move ahead without more resources, and they couldn’t recruit more resources without more progress. Tung, exhausted, was especially anxious and irate.

“We don’t have a singular strategy. We don’t have enzyme. We don’t have crystals. We don’t even have structure activity at this point,” he said. “I’ve been sitting here saying ‘Lord take me now. Let me open my veins.’

“We’re banking that by incorporating some of the changes that other companies have used we’ll be able to improve the overall biological activity of our molecules, but it’s idiosyncratic. It changes from compound to compound. Mixing horses and donkeys doesn’t necessarily get you mules.”

If anything, Tung knew, the decision to go public would now increase the pressure. Immersed so thoroughly in his own work that he had not seen or talked with most of the other scientists in weeks, he was only dimly aware of how the requirements of doing business had come to dominate science elsewhere in the company. But he knew precisely the effect it was going to have on him and Deininger. They were going to have to produce with a late start and a fraction of the number of chemists engaged at Merck and elsewhere what no other company had been able to make: a convincing case that it was close to having a pill to help cure AIDS. To Tung, the very notion was an affront to science.

“We’re in a situation where we can’t afford to make mistakes,” he moaned. “It’s terribly foolish.”

•  •  •

At 2 A.M. Navia, exhausted, again fell asleep in his chair. Boger woke him and told him to go home, then edited the manuscript himself until 5 before driving to Concord, showering, and grabbing a quick breakfast with Amy and the boys. He was back at his desk at 7:30. It was now Thursday morning, the day of the Nature deadline. Murcko, who’d gotten home at 4, was at his computer screen by 8. For a year, he’d tried to be careful about what he ate, but now he was chugging Diet Pepsi and scarfing the gnarled crumbs from a Dunkin’ Donuts box. The room was littered with empty potato chip bags and pizza boxes. Murcko’s briefcase, his floating office, overflowed at his elbow with paperwork—purchase orders, articles he’d been reviewing, candidate’s résumés.

To test his hypothesis about the flap and to begin designing new drugs, Murcko modeled several inhibitors into Yamashita’s finished structure. Boger wandered in, joined by some of the chemists. Collectively, they donned 3-D glasses and speculated on the molecular doorstop theory, whether what they were seeing was real, and what it might suggest.

“It’s reaching in the dark,” John Duffy, a bluff, red-headed chemist, said, looking at a tiny affector region sticking out from the wheel of the protein like a twig from a burl.

“There’s not enough stuff out there to do what Stuart says,” Boger agreed. “Something else has to occur. But our bigger problem is we need a factor of one hundred in activity. We need a break, a bunt single, something as dramatic as the first semaphor.” The semaphor, a two-pronged subunit grafted onto the core of FK-506 like a miniature dousing stick, had been Vertex’s first major success in chemistry, making its molecules as potent as cyclosporine.

The scientists began suggesting several ways to mimic FK-506, but Boger bluntly waved them off. “Nature didn’t design FK too exquisitely,” he said. “This is a molecule that lucked into a use. We shouldn’t be too anal about mimicking it.”

Murcko, taking the cue, disentangled an inhibitor from the protein and spun it around, breaking several crucial atomic bonds. “This is kind of extreme,” he said, looking up from his keyboard. “Don’t try this at home.”

The others laughed. The process of discovery at Vertex, as collaborative as it was meant to be, had a way of shifting the pressure from one scientist to another, like a relay. It was now Murcko’s turn, and he was mindful of how the experience had scalded Thomson, Moore, and Yamashita before him. Boger left, then Duffy, then the others, until just he and Yamashita were left staring into the reflective darkness of their screens.

“Vertex is entirely driven by hubris,” Murcko muttered, half admiringly, half filled with free-floating dread. “We have no reason to expect that we can win, but we refuse to think anything else.”

He was still uneasy about how speculative his work was, more so now that the company was depending heavily on him for its next critical breakthrough, now that its competition was no longer Schreiber but Merck. And yet he needed data. To do serious calculations he needed the coordinates for pinpointing exactly where and how FK-506 bound to FKBP-12—coordinates for which, at the moment, there was only one source: Schreiber and his collaborators. The previous Saturday Murcko had driven to Yale to hear Schreiber. Afterward, Schreiber told him he planned to release the coordinates as soon as his paper was published. Unhesitantly, Murcko called him now. The orgy of posturing, of Schreiber-bashing, was one thing, but this was science. Cordial as ever, Schreiber promised him the information.

Murcko marveled: “Stuart’s really, really good at extracting what he wants from people and dumping them, which is what a lot of people say about Joshua.”

•  •  •

At 11 A.M., after a final strategy session with Boger, Navia, Murcko, and Yamashita, and with the deadline less than an hour away, the X-ray paper was finally done. Moore had taken a couple of calls from the press, but the center of activity had shifted—and held—with the crystallographers. They had made their deadline. Working together, Yamashita and Navia had accomplished what they’d been unable to do apart. The emotional storm of two weeks ago had dissipated, replaced by an uneasy calm.

“Supervising someone else is like sailing a good boat,” Navia reflected. “Everything’s perfectly aligned. The sails are well trimmed, and all you occasionally have to do is gently touch the rudder to steer. With Mason, I’ve had to have my hand on the rudder full-time.”

In another room Yamashita murmered bitterly, “Manuel is a compromised man.”

Yamashita spent the morning agonizing over whom to name as authors. At 11:30, he, Navia, and Murcko wedged themselves uneasily into one of two small private writing rooms off the lunchroom to discuss the matter for the last time. The rooms, barely large enough for a desk, a computer, and a chair, were a concession by Boger to the scientists, a quiet place to think outside the relentlessly public channels of his social experiment. The scientists called them the “domes of science.”

Determined to salvage a shred of idealism from the ruins of his experience with the manuscript, Yamashita had decided to make the decision employing “pure logic, even if we were obliterated by it.” Rationally, that meant naming fewer authors rather than more as he’d previously tried. He now agreed with Boger that the struggle for credit that had erupted so venomously with his original draft of the paper was tearing the company apart and that it was in Vertex’s best interest—in the best interest of science—to downplay such disputes by keeping the number of authors to a bare minimum, perhaps just those two or three who had done the actual work described in the manuscript. In the credit-driven world of academic science, there was an increasing tendency to include all those who, like one’s collaborators, had done other work that had paved the way for a discovery or, like one’s graduate students and postdocs, needed it for their résumés. Boger despised the practice as inflationary and dishonest, and Yamashita now agreed. Coming 180 degrees, he suggested in the dome of silence naming just himself, Navia, and Murcko.

It was a martyr’s position, the most treacherous for Yamashita personally. Thomson and Moore would be—deservedly, Yamashita thought—furious. Moreover, he would look like a credit monger, though that wasn’t his intent. Distraught once again at doing what he had deduced logically to be the right thing, he had already gone to Boger that morning to tell him of the decision, and Boger, anxious to have the scientists put aside their lust for credit, had promised to support him against the inevitable firestorm. Navia, too, now agreed with him. Only Murcko, hoping to lift some of the tension, balked. Telling Yamashita that he hadn’t done enough to warrant credit, he asked him to take his name off of the paper, too.

Thus it was decided. Vertex would set an austere precedent on authorship, striking a blow against fatuous self-interest in research. The X-ray paper would bear just the names of the crystallographers, Yamashita and Navia.

However, what’s logical isn’t always practical, and Boger had designed Vertex above all else to work. Abhorring an anticlimax, he suddenly peeked in, uninvited, yet as if on cue. “I’ve changed my mind,” he said. He told them he now favored crediting all those whose contributions, though not described centrally in the paper, were nonetheless vital. He urged Yamashita to reinstate Thomson, Fitzgibbon, and Moore. Smiling, he ducked out as swiftly and provocatively as he’d entered.

“We were flabbergasted,” Yamashita would recall. “Shocked beyond belief. He said that under the rules of pure logic, it may have been wise for us to confront the problem of authorship now, once and for all. It was probably wiser to sacrifice me rather than to compromise. But he said he understood my pain and Manuel’s pain and Mark’s pain and that the issue wasn’t worth hurting us any more than it already had. He said he’d made a mistake by not letting John publish right away. If he had, he said, the whole question of whether John should be credited on the structure paper would have been a non-issue.”

Boger’s reversal settled the matter, saving Yamashita from all further attempts at self-immolation. At 11:55, Boger faxed the X-ray paper to Nature, feeding the pages into the machine himself. Titled “X-Ray Structure of the Major Binding Protein for the Immunosuppressant FK-506, Solved by Molecular Replacement Using a Structural Fragment Derived from an NMR Study,” it listed as authors Yamashita, Murcko, Boger, Moore, Thomson, Fitzgibbon, and Navia, in that order. Thomson and Moore agreed to be named as coauthors, though in the rush to get the paper out, neither of them saw it before it went. Miffed and annoyed, they refused to share in the overall rejoicing.

•  •  •

“What’s a Judas goat?” Yamashita asked.

He, Navia, and Murcko were still scrambling late that afternoon to assemble photographs and diagrams in order to FedEx them to Nature by 5. The air was aflurry with self-congratulation, forgiveness, the telling back and forth of small lies meant, like an antiseptic, both to sting and heal.

Navia, wolfing a sub and scrambling to make a 6 o’clock flight to St. Louis, where he was due—another of his sidelines—to direct an NIH site visit, explained that it was a goat kept by stockyards to lure other goats to slaughter. A professional traitor, he said. The worst kind.

“Let’s stop playing head games with Mason,” Murcko laughed, pausing. “As long as Mason stops playing head games with us.”

“I don’t want to,” Yamashita said. “How about Judas chickens? Judas cows? Judas . . .”

“Mason is totally awesome,” Murcko said.

“I’ve been saying that since the day he got here,” Navia said.

“His total awesomeness,” Murcko exulted, “is exceeded only by his modesty.”

•  •  •

If Boger’s ability to set improbable goals and then meet them, seemingly against all odds, had not impressed the scientists before, it did now. He had steered the company all week through the sloppy conflation of business and science without so much as getting wet. Now, Friday, was to be his—and their—reward.

The media blitz had worked to perfection. The Wall Street Journal—the only paper that really mattered now, given Vertex’s intention to go public—reported the discovery of the structure in that morning’s paper as an unambiguous three-way tie between Schreiber (in collaboration with Karplus), Clardy (also with Schreiber), and Vertex. “Three teams of scientists reported they had uncovered a key molecular secret of a powerful drug that suppresses the immune system,” the Journal reported prominently on the front page of the second section. Though Schreiber and Karplus were mentioned first, Vertex, being the only business, was identified in boldface, according to the Journal’s standard, busy-executives-don’t-have-time-to-read concession to its readers. Investors looking for an item for their tickler files might not understand the importance of the discovery, but they now had a new name to associate with the “multibillion-dollar market” for new drugs prophesied by Clardy. Of the other papers reporting the story, only the partisan Harvard Gazette failed to mention Vertex prominently.

Schreiber, as Boger had hoped and predicted, was stunned. Recalling his conversation with Moore in late January, which left him, he said, “with the strong sense that they were miles away from having the structure,” he’d also forgotten Moore’s name, the younger, more junior Moore having joined the company after Schreiber became persona non grata and was no longer around. Schreiber had known for weeks that Nature would be reporting a tie, had even heard reference to “Moore et al.,” but had failed to make the connection. “No, I was real surprised,” he would later say, shaking his head.

What the confluence of papers most provided and what was practically forgotten in the stampede to get Vertex’s out the door was affirmation. The ultimate test of whether a molecular structure is right, because it can’t be seen even with the most powerful electron microscope, is whether other researchers, working independently, can duplicate it. That Moore’s and Schreiber’s NMR structures were practically identical and varied only slightly from Clardy’s X-ray structure certified that they were probably both correct. Moore, especially, was relieved: Seeing a preprint of Schrieber’s paper on Thursday, he had exaggeratedly crossed himself. He knew that the differences with Clardy were probably due to the presence of FK-506 in Clardy’s structure. Much more troublesome was the fact that Yamashita’s structure still varied from the other three in several key respects, particularly in the flexibility of the flap. None of the others had the flap swinging so freely. In the general euphoria, the differences between Moore’s and Yamashita’s structures were overlooked, but in private neither of them could release himself entirely from a deep sense of foreboding, of irreconcilability. The idea of a molecular doorstop depended wholly and utterly on the leniency in Yamashita’s structure. Something had to give.

In the scientific community, eyebrows were raised about the latest example of what had become a suspect phenomenon: joint announcements of apparent ties. Under the subhead, “Race to Publish Results,” the Journal noted Nature’s decision to break its news embargo. And in Nature’s News and Views, Dagmar Ringe, a crystallographer at Brandeis University, wrote scathingly that interest in immunophilins had invited a “plague” of publications about them. “More collaboration and less competition in this field would be a damn good thing,” she wrote. For those who worried that biomedicine was being corrupted by blind competition; that in the race to publish, papers were being indiscriminately reviewed or, in certain instances, not reviewed at all; that publicizing discoveries in the media before they had been evaluated by other researchers led to errors, distortion, and fraud; that the need for businesses and academics alike to capitalize on their discoveries created an environment where meaningless redundancy was valued over rigor and innovation, there was now a new case to cite. Not surprisingly, Vertex, as the private entity in the affair, was hoised widely as the black hat.

Boger, of course, had no such qualms. His purposes were clear, and now that he achieved them, he was on to the next siege: going public. He, Aldrich, and Nancy Stuart left Vertex early Friday morning for Warner and Stackpole, the downtown Boston law firm where Ken was a senior partner and which had become a redoubt for Kidder and its lawyers, up from New York. The firm was in one of several overbuilt, underleased, marble-heavy holdovers from the construction boom of the mid- to late 1980s, though the firm itself was prospering. With its ovoid suite of offices encircling a postmodern rotunda, it was the antithesis of Vertex: polished, subdued, conspicuously stylish—old money with a new sheen.

The central question of offering stock in a company is how to value it. With an established business—one that makes money—the process is relatively direct: The underwriter simply mulitiplies earnings by some number reflecting the company’s performance, then divides by the number of shares. With small, unprofitable companies years from having their first product, however, valuation is more mysterious. It depends on an amalgam of assumptions and guesses: When will the company start turning a profit? How much will it make? For how long? What rate of return will investors require, given what they can expect to earn elsewhere? With emerging biomedical companies like Vertex, the questions are particularly elusive. The FDA, not the company, determines when it can market a new product; a company may find out it has nothing to sell only after years of costly development. There are patent infringements, unanticipated shifts in technology, endless questions of product safety. The odds overwhelmingly favor complete, unutterable failure. Still, underwriters try, and the companies do all they can to help them, to put a value on these companies, using formulas and ratios and other more or less empirical measures of things that, like the objects of much scientific study, can’t be seen and may be unknowable.

Boger had long since made his pact with these vagaries, but Kidder now wanted separate confirmation. It wanted to see, as Holman had put it, “if Josh was for real.” Thus the task fell to others: chiefly Nancy Stuart as well as the scientists, who, dressed for the occasion, drove over from Vertex in shifts and cycled in and out of Warner and Stackpole’s conference room like witnesses before a grand jury.

The lone woman in a roomful of men, Stuart led deftly through Vertex’s timelines for making money. The company, she said, expected to identify a new immunosuppressant by the end of the year, launch clinical tests by early 1993, and begin marketing a drug in 1997—some six years away. With HIV, she offered an even more aggressive timeline: identification of a lead molecule by late summer, a drug in clinic the end of 1992, income in 1995. It was a spectacular claim: an AIDS drug ready and available for testing in people in just over eighteen months, profits in less than four years. Stuart tried ably to defend it. Vertex now had novel compounds that were active against HIV in cells, she told the lawyers. It had contracted with a California company to test them by midsummer in a new breed of genetically engineered laboratory mice that had most of the components of a human immune system and were infected with HIV. If the compounds worked, Vertex would be able to short-circuit the normal preclinical schedule and go directly into primates. Like Merck, it would then launch European clinical trials while compiling data for the FDA, which was desperate for new AIDS drugs and was pushing ahead furiously anything that looked the least bit promising.

Stuart worried that the Kidder men would consider the story too speculative, too filled with holes, too pat. “People have heard this a hundred times before: new technology, bright people. ‘Yeah, yeah,’ ” she interrupted herself, speaking afterward, “ ‘so tell us about your clinical candidate.’ ” But Kidder was impressed. The biggest worry about small biomedical companies was that they would go on hemorrhaging money forever, but Vertex was obviously determined to make money soon. Indeed, it was the main reason it had ultimately gone into AIDS. True, dozens of other companies were fast-tracking potential AIDS drugs for entirely the same reason, but once again Holman found Vertex’s story exceptional.

The scientists, on the other hand, were haunted. To make public their aggressive timelines simply added to their stress. “The HIV project is stretched outrageously thin,” Tung complained. “What concerns me is that even in the most optimistic scenario, we may not be able to do what we’ve said we’re going to do. It would be hard enough if we were fully staffed, but I don’t see that happening now. We’re running a lot more on luck than skill and that worries me. It worries me a lot.”

Murcko, who also was interviewed, was even more outraged. He was disgusted by the intrusion. Yanking off his tie as he returned to the modeling room, he snapped: “A $500 lunch for fifteen people, and they billed us for every Coke we ordered.”

•  •  •

Back and forth between the real world of scientific speculation and the illusory one of Wall Street Boger and the scientists swung. At noon, while Boger, Murcko, and Navia were lunching at Warner and Stackpole, a fax arrived from Nature. It said that in light of the similar X-ray structure of FKBP-12 appearing in that day’s issue of Science, the journal had decided not to send Vertex’s manuscript out for review. Yamashita, reading it, went cold. Twenty-four hours earlier, he, Navia, Murcko, and Boger had exulted not only about catching Schreiber, but about staking out an important new hypothesis about the biology of the world’s hottest protein. They had driven themselves recklessly to meet a deadline that served Nature as much as it did them. Yet now their work wouldn’t even be seen by those in a position to judge its scientific merit. Again, as it first had with Moore, Nature seemed to be rejecting a potentially vital piece of science for no reason but its own competition with a rival journal.

Boger took the letter immediately on his return and began composing his reply, an artful, exaggeratedly civil single-spaced three-page rant. Yet another occasion for Boger to rise to after a week of nearly continuous crises, it showed him at his most brilliant and most caustic.

Scientifically, he wrote, Nature had “compelling” reasons to reevaluate its decision. First, there was the significance of the structure itself: It was unique. Unlike Clardy’s structure, it was unbound (it was the first X-ray structure of the enzyme by itself), yet it was distinct enough from both Moore’s and Schreiber’s NMR structures to raise the question of what the protein really looked like. At the same time, it posed a direct challenge to Clardy and Schreiber, who had observed that when the protein bound to FK-506 its overall conformation “changes little” from its native state. “As you will note from our data,” Boger wrote, “this would appear to be obviously and significantly incorrect. . . . Rather significant changes in protein structure must occur [emphasis added] on binding FK-506, especially in the ‘flap’ region that guards the binding site.” He also stressed the collaboration between Yamashita and Moore—“only the second case of NMR data being used to solve an X-ray structure” and “the first time” a fraction of an NMR structure had solved the problem of molecular replacement.

Any of these points might have been enough to warrant Nature’s reconsideration: Taken together, they made a convincing case. But Boger couldn’t resist going further. He bluntly reminded Nature’s editors of the deal they had made in deciding to reconsider Moore’s NMR structure. “We would not be frank if we did not express our disappointment at your abrupt reaction to our manuscript,” he wrote near the end. “The favorable editorial decision in mid-April to send the NMR manuscript for review was made only after, at your request, we laid out in detail our X-ray observations and plans and committed ourselves to sending the as-detailed manuscript to you. We fully and completely met our side of what we cannot help but view as something akin to a bargain.”

Boger had not come this far to be denied. He concluded the letter by giving Nature until the end of business Monday to reverse itself—one day. Otherwise, he said, Vertex would take its side of the FK-506/FKBP-12 story—“perhaps the ‘hottest’ subject in medically relevant biology,” he reminded scoldingly—to a competing journal.

To the scientists, including Navia and Yamashita, whose names Boger affixed to the letter above his own, the letter was brash, impolitic, shocking, and potentially ruinous. They feared Nature would never again publish a manuscript with the company’s name on it. Yet by now they had no choice but to go along. “We’ve already torched rule 1—‘Never publish a paper we’ve already rejected,’ ” Navia said gamely. “There should be no problem torching rule 62—‘We don’t publish second X-ray structures.’ ” Still, underneath lay a jagged ribbon of fear. All they could do was wait and hope that Boger’s hubris wouldn’t lead them into an all-out Icarus-like plunge.

There was a five-hour time difference between London and Cambridge. Nature had until noon Monday, local time, to respond.

•  •  •

Nature’s second reversal in a month came exactly twenty-seven minutes late, at 12:27. Boger accepted it munificently. It still meant that others—others more rigorous, closer to the situation—would have to approve the X-ray paper, but to Navia, the last great obstacle was now cleared. The stage was set for a full-scale confrontation with Schreiber—a confrontation he was every bit as confident as Boger that Vertex could win.

“Stuart has stuck his neck out,” Navia said. “His work is a hypothesis driving a conclusion. He thinks FKBP is an inconvenient intervener in the process, that it’s a socket for FK-506. But he doesn’t understand biology. The Nature paper is going to rough him up a bit.

“If we get the complex,” he said, almost as an afterthought, “we’re going to chop his balls off.”