Here Is a Human Being

by Misha Angrist

  But if I were expecting an angry, militant black dude, I was to be disappointed. James spoke with passion about his cause, regret about what the ordeal had done to his family and friends, and anguish about the time it had taken from his science. He also laughed easily and told me he was relishing the time at home with his kids while he looked for a new job.138

  When we met in a Brookline coffee shop the following spring, he was less preoccupied, though he assured me he was not at peace. The Equal Employment Opportunity Commission, with whom he had filed a discrimination suit, had not found in his favor;139 this discouraged him. And he was still angry with some of his allies, who he felt did not call out the MIT provost with sufficient vigor. He blamed himself for the same lack of will: by not resuming his hunger strike, he said, he felt he had let MIT off the hook (although he admitted that his decision also spared his wife and eleven-year-old daughter further anguish; his daughter was especially concerned that her father really would “die defiantly”).140

  By the second PGP gathering his professional life had begun to settle. He was now a senior scientist at Boston Biomedical Research Institute in Watertown. The director was already familiar with James’s science. But did BBRI flinch at his stance on ESCs or his MIT baggage? “There was one question during my interview: ‘How do you feel about ESC research at BBRI?’ My response was truthful: I’m not going to get in the way of any programs that involve ESC. But I’m also not going to participate in them. I will continue to be an outspoken advocate on this topic, but I’ve never interfered with anybody’s research.”141 Did James Sherley deserve tenure? I have no idea. Most junior faculty at MIT don’t get tenure; this, after all, is why it’s MIT, a truly elite institution. Had he won the Pioneer Award two years earlier, before the decision and the subsequent escalation, perhaps things might have gone differently for James. And it’s not hard to see how he might have hurt his own cause with his demonstrativeness, his high-profile martyrdom, and his pointed and often public accusations. But my guess is we will never know because I suspect the process was doomed from the beginning by miscommunication and a lack of transparency on the university’s part.

  In January 2010, MIT published a comprehensive study of racial diversity among its faculty142 (disclosure: my brother Josh is an economist at MIT and was on the committee that crunched the numbers for this report). While the Sherley incident was mentioned just once in the 156-page report, I found it hard to believe it was not a major impetus for MIT’s decision to reflect upon its own minority hiring and promotion track record. Among the findings: 74 percent of Caucasian assistant professors were promoted to associate professor without tenure (a crucial step on the path to tenure), while only 55 percent of black and Hispanic assistant professors were promoted to associate—a statistically significant difference. Of course, only 6 percent of MIT’s faculty was black or Hispanic to begin with (a fraction comparable to Harvard and Stanford).

  Perhaps future African American scientists—and those who make the tenure decisions affecting them—will benefit both from the report and from the Sherley episode itself. That might be the best we could hope for.

  After the tenure drama and the hunger strike and the protests, George hesitated to ask James to participate in the PGP. Not because he shied away from controversy (he would not be George Church if he did that), but because he couldn’t imagine his friend and collaborator would want to undertake yet another high-risk, high-profile controversial venture. He was wrong.

  “I kept wondering, ‘Why isn’t George asking me about this?’ “ said James. “When he finally did I was like, ‘What took you so long?’ I’m interested in being fair in disclosure and thinking about how to build research programs that can actually be responsive to the needs of participants in the study.”143

  Eventually James and I got around to discussing our genomes, and the aggrieved academic instantly morphed into the awestruck nerd. James was not terribly interested in genes per se (cell biologists often operate at a higher, more systemic altitude than molecular biologists).144 At one point during the initial PGP meeting he said he doubted he would even look at his own genome, though he later backed off from that statement. But he said he was much more curious about his “origins of replication,” the sequences where DNA initiates the process by which it copies itself; there is some controversy about where they can be found in the genome.145

  James’s real passion was stem cells. Stem cells are undifferentiated cells. We all begin existence as embryonic stem cells. They are valuable because their fate has not yet been determined; they have not committed to becoming neurons or T cells or muscle cells or melanocytes. Their value in medicine, so far largely unrealized outside of leukemias, is that they have the potential to become almost any type of cell (for this reason they are referred to as pluripotent). They could become brain cells in a Parkinson’s patient or pancreatic islet cells in a diabetic. If we want to get the greatest number of stem cells with the broadest range of possible fates, embryonic stem cells are ideal. But many, like James, are not comfortable with destroying them because they see them as human life. These scientists have therefore focused on reprogramming adult cells: taking a skin cell, for example, and “dedifferentiating” it back to a stem cell state where anything, or many things, are still possible. This is called induced pluripotency: taking a differentiated cell and turning the clock back so that it behaves like a stem cell.146 Sherley’s lab had helped figure out how to do this and how to get these cells to divide in culture so that they generated a renewable supply of themselves.147 The problem was that so far it was a terribly inefficient process. Making it more efficient was one of his and George’s goals.

  James was excited about stem cells for other reasons. One was the biology they might teach us. Another was their potential predictive value. “Cancers may be stem cell diseases. Asthma, too. Multiple sclerosis may be a stem cell disease. Changes in the genome of undifferentiated cells may predict [health outcomes] because those are the cells that are constantly renewing.”148 By the time we reach the fetal stage (the ninth week of pregnancy), our developmental potential has been written into most of our cells. “If you could only make one gene expression pattern in a person’s life and see which genes are turned on and off, the fetal period would be the time to do it.”149

  Different genes are turned on and off at different levels at different times in different types of differentiated cells; this is reflected in the different levels of RNA one finds for the same gene in different cell types at different stages of human development. This is what makes a neuron a neuron and not a hair follicle. James and George wanted to conduct an “organ recital"—that is, they wanted to measure gene expression in each of the 200+ different types of cells in the human body.150 For blood, hair, and umbilical cord, that was pretty easy. Buccal cells could be scraped from the inside of the cheek with no problems. Skin biopsies hurt but were tolerable. But once you got into muscle and liver biopsies, to say nothing of brain tissue, then access became a real issue. So why even bother? Because if we really wanted to understand human beings at the molecular level, organ recitals were a necessity. “We’re more than our genes,” explained James. “We’re the expression of our genes. Looking at differences in expression is going to be much more informative than genotype.”151

  It was 1:30 p.m.; we had been sitting in the coffee shop for more than two hours. Outside we could still see our breath; we had to step over the snowbanks to cross the street but the sun was shining brightly in Brookline. As we walked to James’s car, I asked whether he thought the PGP could really make a difference. Wasn’t personal genomics a luxury? “Listen, there are lots of other needs in society. Should we be spending this money on building houses for people? I don’t know. I certainly think and worry about it. But as a scientist I feel that knowledge shared is the best thing we can have. What should we be gathering knowledge about? Understanding how we work, how we function, how we grow up and smile—that will be difficult to do. But it’s hard t
o argue that that isn’t important somehow.”152

  * Pharmacogenomics is the science of using an individual’s genetics to determine which drug and what dose will be most effective. Medco is a Pharmacy Benefit Manager: it manages the prescription drug (and sometimes other) part of an employer’s health benefits. PBMs try to control administrative costs of processing prescription drug insurance claims and ensuring that beneficiaries are taking the right medications for the right conditions. One can see how genetics might help a PBM by pointing patients to drugs that are more likely to work based on their particular drug-metabolizing gene variants.

  * This did not stop 23andMe from returning results on a subset of patented breast cancer mutations, however; the company began offering these results in February 2009; http://www.genomeweb.com/dxpgx/23andme-adds-brca-breastovarian-cancer-testing-service.

  * While our genomes are 99.9 percent identical to one another on average, the 0.1 percent still adds up to 3 million differences. And because human populations were geographically separate for long periods of time, it is possible to make fairly precise estimates of someone’s continent of origin based on a few thousand DNA markers. Indeed, this is how 23andMe (see chapter 5) and companies like Family Tree DNA assess their customers’ origins.

  * MIT faced a minor insurrection by female faculty members in the 1990s, which resulted in an acknowledgment of gender discrimination and a concerted effort to address it.

  * The provost by this time was L. Rafael Reif.

  7 “It’s Tough to Guard Against the Future”

  Thanks to the late, great, and supercheap Skybus Airlines, Ann and I could afford to bring the kids to what had become, after only one visit, my favorite meeting: Advances in Genome Biology and Technology, in Marco Island, Florida. As I intimated in chapter 5, AGBT is to DNA sequencing what Macworld is to all things Apple: a highly anticipated annual unveiling of sleek new toys, replete with the requisite back-channel discussions, competitiveness, fire hoses of data, and more than a little showmanship, all at the plush Marriott resort right on the beach.

  AGBT could be counted on for theatrics (Pacific Biosciences, still more than two years away from launch, sponsored fireworks over the Gulf) and rumor-mongering (“So and so’s new machine is having problems”). But by 2008, the next-generation sequencing field had gotten more crowded,1 with Illumina having overtaken first-to-market 454 and ABI trying to play catch up after launching only a few months prior to the February meeting.2 Half a dozen other companies claimed to be “close”

  to bringing instruments online. Attendees heard big-picture, crystal-ball talks about “the future of DNA sequencing” and more narrowly focused presentations with sexy titles like “Respiratory Bacterial Pathogens Utilize Polyclonal Infections and a Distributed Genome as Population-Based Chronic Virulence Traits.” I confess I skipped that one for some quality time in the Tiki Pool and a few drinks with umbrellas in them.

  George was not there in the flesh, but certainly was in spirit. On a late afternoon inside a rented suite at the Marriott stood his lab’s crowning technological creation (at least for the moment): the Polonator. I had already seen the hastily assembled marketing piece, which was adorned with a picture of a large blue box that had been slightly warped in kind of a Daliesque way, a nice design touch and appropriately Churchian. The splashy-cum-nerdy brochure announced the machine’s arrival:

  Dover, in collaboration with the Church laboratory of Harvard Medical School, introduces the Polonator G.007, a revolutionary approach to second-generation sequencing. The Polonator G.007 is a completely open platform, combining a high performance instrument at a very low price point … [Users] are totally free to innovate; all aspects of the system are open and programmable… .

  [The Church lab’s] vision, as expressed in the Personal Genome Project, the development of the Polonator, and their recent formation of the PGx team [that will compete for the X Prize], is quite simple: to deliver the benefits of second-generation sequencing to the largest possible base of potential users, as quickly and efficiently as possible.

  … For those intrepid souls willing to slip into the driver’s seat, the Polonator is completely open and at your disposal. Beyond buckling up, our only request is that you respect the open nature of the Polonator system, and promptly publish (or better yet communicate immediately via our user community forums) any enhancements that you might develop. It is through your creativity that the Polonator system will evolve.3

  The Polonator would be more than just a sequencer; it would be a philosophy—a way of life. Like the PGP, this thing was not for the timid. Why this approach, George? “For anything you do on government grants,” he said, “open-source is a good idea. Also, I survived the Applied Biosystems monopoly; I always felt it was too hidden and it was too stifling. I always felt like, ‘Gee, if somebody had the resources, wouldn’t it be great to just have all this stuff in the open?’ I think that feeling comes from having benefited from open-source software myself. And it also seems to be consistent with what we’re doing with respect to the ELSI [ethical, legal, and social issues] aspects of the Personal Genome Project. We are trying to be transparent in every way.”4

  Applied Biosystems was the company that won the race to develop automated sequencing in the 1990s. By convincing bad-boy scientist Craig Venter to undertake a private effort to sequence the human genome and compete with the government’s Human Genome Project, ABI became the arms dealer that catered to both sides of the war. Venter and the NIH-sponsored public effort each bought hundreds of ABI’s instruments at three hundred thousand dollars a pop. And with the company’s model 3700 (and its successor, the 3730) installed as the industry standard, ABI could then charge lots of money for reagents: proprietary molecular bullets for the company’s high-tech guns. It was a brilliant move and it helped to keep ABI’s balance sheet deep in the black and its machines entrenched in hundreds of molecular genetics labs for the better part of a decade.5 But now obsolescence was in sight: the wealthier labs had already begun to switch to the new platforms and draw down sequencing on the old machines.6 In 2009, I saw a used 3700 on eBay for about $3,500, shipped. In another auction, its predecessor, the 377, was sold for a winning bid of $99.

  In the suite at the Marriott I helped myself to a glass of red wine and inspected the Polonator up close. Even someone who had no idea what it was had to be impressed with the aesthetics. It was electric blue, a much bolder color than any of the competition’s wares. Its glass door was flanked above and below by bands of orange racing stripes. Inside were all of the usual moving parts characteristic of a next-gen sequencer: a robotic platform to cradle the flow cell—that is, the small piece of glass that held the DNA to be sequenced; a charge-coupled device camera to record the images of each base after it was incorporated; and lots of tiny capillary tubes that moved enzymes and other reagents in and out of the flow cell. Next to it was a computer whose job it was to instruct the Polonator. And lo and behold, this Polonator was actually on! Its lights were flashing and its parts humming as they moved stuff from one place to the other. If this thing worked, I thought, I might actually get my sequence someday.

  The man who had taken the technology from the Church lab and turned it into an exemplar of sleek design was Kevin McCarthy. Were one to draft a prototypical George colleague from the technology sector, Kevin would probably be pretty close to the final iteration: a tousled mop of gray hair, skinny with a buttoned-down shirt, oversize wire-rimmed glasses. An engineer not entirely comfortable as pitchman but full of ardent belief in his product. An inventor.

  A year earlier, McCarthy, who’d been put on to the Church lab by a sales rep, was being escorted to Harvard Medical School to meet with George’s team. But his coworker/driver got lost in the medical school maze. By the time they made it to George’s office, they had all of fifteen minutes to make their pitch. “And it wasn’t like they were dying to see us,” McCarthy said.7

  At the time the Polonator was much closer to a science fai
r project than a commercial product. When I first visited, “Polonator Central” was still a small windowless room within the sprawling Church lab. Inside a boom box throbbed with the Rolling Stones’ Steel Wheels. The first thing one noticed was the temperature: 18° C, or about 64° F. Ligase, the enzyme used to stitch DNA together in the “Polonation” process, liked it cool; early versions of the Polonator had no onboard refrigeration (or onboard much of anything), so Church protégés Jay Shendure and Greg

  Porreca had to keep the ambient temperature down. The room was littered with tangles of black wires connecting microscopes to computers on stainless steel shelves. Space was so tight that the keyboards dangled vertically over the sides; Rube Goldberg would have been proud. Each one of these makeshift arrangements had been named after a character from The Simpsons. Marge, I noticed, was in the midst of running a batch of samples. On the screens were thousands of white dots, like stars on an especially clear night. These were polonies—short stretches of DNA that had been amplified millions of times.

  “I wandered into the lab and saw some interesting things,” McCarthy recalled diplomatically. “I said, ‘I can do a lot of these things better than this stuff you’ve cobbled together.’ It began with a few core components and just kept getting bigger. I said, ‘Would you like some metal to connect all this stuff?’ Finally it got to the level of, ‘How about we just make the whole thing?’ They were like, ‘Yeah! Let’s go!’”8

  Back in the main ballroom, Baylor’s Amy McGuire was making a presentation on the ethical challenges of personal genomics. We were living in an exciting time, she said, but the excitement was tempered by worry. All of the events of the last year or two—the publication of individual genomes, the X Prize, NIH grants meant to hasten the arrival of the thousand-dollar genome, genome-wide association studies, and yes, the PGP—raised some ethical issues that for the most part had not been in play before the digital and genomic ages. Among the biggies: