What will the world be like when your genome sequence costs less than a cell phone? A couple days ago I went to Cambridge, Mass. to find out.
The occasion was a meeting called “Genome, Environments, and Traits,” or GET for short. The history of the meeting is in the upper ranks of my list of meetings with strange histories. In 2006, the Harvard geneticist George Church (arguably the smartest, most influential biologist you never heard of) decided to launch a new kind of human genome project. At the time, scientists had only published the sequence of a single human genome, at a cost of $3 billion. And for all that money, the genome was actually a gap-riddled patchwork from several individuals, and only included the DNA from one copy of each pair of chromosomes. Church declared that he would gather the sequenced genomes of 100,000 individuals, along with information about their health, and make all that information available for scientists to study in order to learn more about human biology. Church issued a manifesto of sorts in Scientific American, called “Genomes for All,” which you can read here (pdf) and also talked to Emily Singer of Technology Review here.
To kick off his Personal Genome Project, Church sequenced his own DNA, put it online, and promptly got a message from a doctor on the other side of the country, informing him that he should adjust his cholersterol medication. Church also persuaded nine other people to volunteer to have their genomes sequenced and laid out online for all to see. One of those first ten, the Harvard psychologist Steven Pinker, helped spread the word with this article that in the New York Times Magazine in 2009.
Those early sequencees got together from time to time to talk about the project and their own experiences contending with having a genome available for all to see. This genomic club was an intimate one at first, but its membership is now exploding. With each passing month, the cost of genome sequencing is crashing, companies are gearing up to sequence genomes on a commerical scale, and scientists are starting to think seriously about looking at complete genomes as a regular part of clinical practice. For this year’s meeting, Church decided to try to get as many people with sequenced genomes as possible altogether in one room. It would probably be the last time such an exercise would be possible.
I got pulled into the fun when my phone rang a couple months ago. On the line was Robert Krulwich. Krulwich is the co-host of the show Radiolab, covers science for NPR and ABC News, and is also the go-to guy for live–and lively–interviews with towering figures of science. (Observe him handle both E.O. Wilson and James Watson at once–a bit like juggling torches while riding a unicycle. He doesn’t break a sweat.)
Church had asked Krulwich to come to the meeting and moderate a discussion of a dozen or so sequencees that would take up the first three hours of the meeting. Krulwich decided this was a two-person operation. Wise move. This was heavy lifting.
It would be absurd to have everyone one the stage the whole time, so we came up with a scheme to move people quickly from the front row of the audience to the stage, playing a genomic game of musical chairs. Making it even more challenging was the fact that we had such big subjects to talk about, from the development of next-generation sequencing to the application of genomics to genealogy to the issues of privacy that genome sequencing raises.
And then there was the matter of the line-up. Any one of the speakers could have held the stage on his or her own for an hour. It felt very strange whisking Henry Louis Gates onto the stage and then whisking him off. This, after all, is a guy who can hold an entire TV series together. At the meeting, he talked about getting his father’s genome sequenced as well as his own–becoming the first father-son team to do so. A comparison of the two genomes allowed him to see fifty percent of the genome of his deceased mother–an experience that felt like seeing her come back to life. Gates talked about the experience of seeing so much European DNA in his genome. If you look at my lab results, he said, I’m a white man.
–Well, we’d love to hear all about it, Professor Gates, but we’ve got to move on! A round of applause everyone, and let’s move those chairs!
Krulwich and I also struggled with the challenge of talking about genomics with people who are so uniformly gung-ho about it that they’ve had their genomes sequenced–and of talking to those sequencees in front of an audience made up of genome scientists, people from the biotech sector, venture capital folks, and other assorted people who are, shall we say, already in the genomic tank. Neither Krulwich or I received a fee for our involvement in the meeting, and we were not about to join the ranks those miserable fake journalists you see on infomercials late at night, pitching pre-scripted softballs like, “So tell me again how your company is going to become a raging success in the personal genome business.”
Krulwich and I therefore tried, politely, to nudge the sequencees out of their comfort zone. How on Earth, I wondered, could the sophisticated analysis of genomes become a regular part of everyday medicine when most doctors have office full of old paper records? Was it fair to children to get their genomes sequences when there was nothing immediately wrong with them? What good is getting your genome sequenced if all you get is a laundry list of genetic variations that have obscure relationships to all sorts of diseases that you may or may not get? How can there be a business in genomes if, as Church predicts, the cost of genome sequencing will be dropping to, essentially, free?
In many cases, questioners and answerers ended up talking past each other. Krulwich asked James Watson what he thought about the ethical concerns about genome sequencing. His answer: “Crap.” The other sequencees were more polite when we asked questions that seemed irrelevant to them. When Krulwich asked sequencee Esther Dyson about the potential risks of getting her genome sequenced, Novocell CEO John West pointed out that she was preparing to go to the Space Station. Why were we obsessing about the risks of Dyson’s genome, with no apparent concern that she was about to have herself shot into space on the tip of a rocket?
I think the best answers were deconstructions. Consider this: Widespread genome sequencing will make it possible to test babies for genes associated with intelligence. Isn’t that a horrible thing?
At the meeting, Church pointed out that we already test for intelligence genes, and nobody gets outraged at all. Babies are routinely tested for a genetic disorder known as PKU, in which children are born unable to break down an amino acid called phenylalanine. Phenylalanine builds up to toxic levels in the body, leading to mental retardation. But the mutation that causes PKU does not necessarily cause PKU. Genes are not destiny. If children keep a diet low in phenylaline, they end up with normal intelligence. Knowledge of our genome is not sinister in this case. Ignoring the facts of PKU would be the sinister thing to do.
Church is right, but the story of PKU only carries you so far into the future of genomic medicine. PKU is a rare disorder, affecting an estimated 1 child out of every 13,500 to 19,000 births. It’s also unusual in that it’s caused by the failure of a single enzyme. A single mutation to a single gene is enough to cause it. And the fact that it can be so readily treated is also unusual. Cystic fibrosis, for example, is another single-gene disease. Despite the discovery of its genetic basis 20 years ago, doctors have no cure to offer CF patients.
The genetic roots of common disorders, like high blood pressure and Alzheimer’s disease, have proven to be a lot more complex. It’s possible that the risk for some common diseases may be the result of variations on hundreds of genes, with each variation contributing a tiny fraction of the risk, and different combinations able to cause just as much of the disease. It’s also possible that the risk for some diseases is due to very rare mutations, each of which has very strong effects. There may be a lot of these rare mutations in the world’s population, making it hard to find them all and figure out what they do.
The sequencees at GET didn’t avoid this messy reality. In fact, one of them embodied it. James Lupski, a Baylor College of Medicine geneticist, suffers from a hereditary disease called Charcot Marie Tooth Disease, in which the coating of the long nerves in the limbs starts to fray. He has had to have operation after operation on his feet to treat the symptoms. Lupski studies the cause of the disease, and recently he had his genome sequenced to find its source. He turned out to have some mutations that have been linked to Charcot Marie Tooth Disease before, but he and his colleagues also found a new gene, with a different mutation in his mother’s and father’s copy. The discovery did not point immediately to a cure; instead, it added to the complexity of the disease. Lupski explained his own disease and his difficult research on it in unsentimental detail. Science is hard, Lupski said, and anybody who thinks it isn’t is fooling themselves.
It was too bad that the meeting didn’t take place next week instead of this week. Today, the Lancet published a genome paper that included among its co-authors two of the sequencees we spoke with: George Church and Steven Quake of Stanford. At the meeting, Quake explained how he and his colleagues had sequenced his genome last year in a matter of days. That was the easy part, he said. The hard part was analyzing it and interpreting what it meant for Quake’s health. He was referring obliquely to the new paper.
In the paper, Quake, Church, and their colleagues made a close study of Quake’s family (who have suffered from various sorts of heart disease), and then scoured the scientific literature for every mention of the variants they found in Quake’s genome. They considered the risks these variants posed to Quake for various conditions, but they also took into consideration other sorts of complexity. For example, diseases don’t happen in isolation from each other. If you get obese, for example, you increase your risk of type 2 diabetes. The scientists published a marvelous diagram of the diseases they studied in Quake, with the size of each name corresponding to the size of his risk for each.
The geneticist Daniel Macarthur wrote tonight about this new paper on his blog Genetic Future:
…there are the variants that simply can’t be interpreted. This includes virtually everything seen outside protein-coding regions, and the majority of even those variants found inside coding regions. We simply don’t understand the biology of most genes well enough yet to be able to predict with confidence whether a novel variant will have a major impact on how that gene operates; and we have an even less complete picture of how genes work together to affect the risk of disease.
Like Lupski said, science is hard.
I was wiped out by the end of the morning session. I thought we did a pretty good job, although I still felt ambivalent. I scarfed some lunch and then happily settled into the audience for the afternoon. Most of the talks I heard dealt not with humans but with microbes. The genome of a microbe like E. coli is about a thousandth the size of a human genome. As a result, microbiologists can sequence genomes like mad without busting their budgets. Ian Lipkin of Columbia has hunted for the causes of new outbreaks, such as colony collapse disorder in bees, by fishing out new kinds of microbial DNA from sick hosts. Boom, boom, boom, one slide after another documented the discovery of yet another pathogen. The benefits of DNA sequencing were blindingly obvious in Lipkin’s talk.
But even microbes turn out to have fantastic genomic complexity. There may not be a lot of genes in each microbe, but together they can hold a staggering amount of genetic diversity. Rob Knight of the University of Colorado spoke about the surveys he and his colleagues have made of the human microbiome. He described some of the work I’ve blogged about here on the Loom, along with other results. He described, for example, how children become coated with the bacteria that live in their mother’s birth canal as they are born. Women who have a caesarian section give their children the bacteria living on their own skin. Knight is investigating whether the birth canal germs provide any special protection to children. Different people develop different menageries of microbes as they get older, and their experiences–from gaining weight to taking antibiotics–can shift the ecosystem inside their bodies. There’s much left to discover about the thousands of species that share our bodies with us, but Knight raised the prospect of a different kind of personalized medicine: using genomics to survey the microbes in our bodies and then manipulating them for our own benefit.
Then again, maybe you shouldn’t trust me on this score. Everyone knows I’m in the microbial tank.
The day ended with a talk by Anne West, the 17-year-old daughter of John West. The Times of London recently broke the story of how the Wests became the first healthy family to get their genome sequenced. I expected warm and fuzzy blather about what her genome meant to her, but instead, she delivered a hard-core talk that would have fit right into a genetics conference. She analyzed one of her genes involved in blood clotting and determined that she had a few harmless mutations from her mother and one harmful one from her father. Facing an audience full of past and future Nobel-prize winners, biotech barons, and other intimidating grown-ups, she remained impressively poised and calm.
The audience was rightly impressed. One scientist joked that she should drop out of 11th grade and get a job–finishing school would be a waste of her time. But I also had to remind myself of the hothouse atmosphere in which she had done this work, and in which she was delivering her results. Her father had spent upwards of $200,000 on the family’s genomes, according to the Times. This was not your standard science fair project. And as West spoke, I thought about the kids from a local public high school who had come for the morning session. When Krulwich and I asked the audience for questions, a girl stood up and asked how she could get her mother to have their family’s genomes sequenced, when her mother wasn’t even sure what a gene is. Two girls: two very different experiences with genomes. It’s not all about the DNA.
[PS–Thanks to all the Twitterers who acted as a note-taking collective. Their assembled chronicle is here.]