When a first draft of the human genome was sequenced in 2001, biology suddenly became much bigger. The genome—three billion DNA letters long—contains an incredible amount of data. But until recently, medical biology had been concerned with drilling down to the finest details of how the body works. With genomics, on the other hand, it is now possible to get a wide-angle view of human biology. Fifteen years ago, we thought of the human genome as this infinite, unbounded territory, says Eric Lander, a faculty member at MIT who is also professor of systems biology at Harvard Medical School and head of the Broad Institute at Harvard and MIT. The Human Genome Project rendered it finite, just big. So now you can ask questions like, What are all the genes that confer risk for diabetes? What are all the mutations that might give rise to cancer? What are all the genes that are essential to make a cell infectable by HIV?
The Broad (the name rhymes with road) is a joint venture among Harvard, MIT, the Whitehead Institute for Biomedical Research (also located in Cambridge), and the Harvard-affiliated hospitals that aims to answer some of those big-picture questions. We have a really clear mission, Lander says. We want to fulfill the power of genomics for medicine. Announced in 2003 and officially launched in 2004, the Broad focuses on interpreting the information in the genome in a way that can be applied to human health. It also serves as a kind of service center for scientists at the two schools who want to apply the tools of genomics and other large-scale technologies to their own work.
Making the genome useful is a much bigger task than sequencing it. The Human Genome Project produced a single DNA sequence, but in reality there is no one human genome. For genomics to be useful to medicine, it must tackle the differences among people that lead one person to get a disease and another to avoid it. David Altshuler, associate professor of genetics, one of the core members whose labs are based primarily at the Broad, leads the medical and population genetics program, which studies such genetic differences. One of its key goals is to understand the genetic basis of complex diseases like diabetes and heart disease. Altshuler, who trained as a diabetes clinician, became frustrated by the lack of knowledge about the diseases origins. So many of my patients had parents and siblings and children with type II diabetes, and we couldnt explain it at all, he says. And it seemed to me amazing that in this era of genetic information, we didnt know what was causing it.
Standard biology has been very good at explaining the so-called Mendelian genetic diseases, in which inheriting a mutated gene leads something in the body to go awry. Complex diseases, however, seem to be a messy combination of genes and environment. No one gene causes a disease like diabetes, but several genes might put a person at higher risk for developing it in the presence of certain behaviors or environmental factors. To tease out risk genes, researchers collect DNA samples and clinical information from large groups of people with and without the disease and look for statistical differences in their genes that may reveal how and why the disease occurs and may also give scientists a guide for developing treatments.
Psychiatric diseases, very few of which have any known cause or mechanism, are perfect candidates for genetic analysis. Edward Scolnick, senior lecturer on genetics, the former president of Merck Research Laboratories, has changed direction in his career to address this problem. Now, as director of the Broads psychiatric disease initiative, his goal is to change the way such diseases are treated. The effort focuses on schizophrenia and bipolar disorder, which seem the most likely to be heritable. Scolnick says that knowledge of the biological basis of psychiatric diseases is woefully lacking: There is no chemical test or physical test or biological test that allows you to make a diagnosis of these diseases. The best way to get at the biology is through genetic studies, he says. To really understand function you have to understand how the genes affect that function.
Genetic studies on this scale have become feasible only in the past year, and the first papers are just being published. It is likely that genomics will quickly shed light on some diseases, while others may remain a muddle of data. But the Broad leaders insist that, whatever the outcome, science will benefit from a wide-angle view of the relationship between genes and disease.
An ambitious application of that principle is the cancer genomics group, led by Broad core member Todd Golub, associate professor of pediatrics. The group uses genomics to study and classify different types of cancers. This is not as simple as it sounds: part of what defines a cancer cell is a genome that is irregular and unstable. Not all cancer researchers believe it will be easy to make sense of erratic cancer genomes, but Golub says that its important to try. Most cancer research involves finding a gene involved in some aspect of cancer and studying it in great depth, he says. Entire careers are spent cataloguing every aspect of a genes function. If thats the only approach you take, theres the possibility you dont get the big picture. You see only the bits you know about; there may be vast pieces of information that remain.
By bringing together an impressive array of technologies under one roof, the Broad has created a new kind of resource for Bostons biomedical community. Lander likens it to NASAs Jet Propulsion Laboratory in California, based at the California Institute of Technology. A grad student at Caltech may come up with an idea for sending something to Mars, but when it comes time to build the rocket, she relies on the expertise of the laboratory. We havent got a tradition of that in biology, Lander says. Thats what the Broad is. We are trying to put jet rockets under smart young faculty, postdocs, and grad students.
Until now, biology simply hasnt had many tools equivalent to jet rockets. But in the past several years, the technology available to biologists has become more powerfuland expensive. Genomic sequencers are the resource most people associate with this kind of research, but other technologies also help to answer questions on a large scale: DNA chips can report which genes are turned on or off in a particular cell; RNA interference allows scientists to turn off selectively each of thousands of genes in a cell; and libraries of chemicals let researchers quickly screen many different molecules to identify potential candidates for future drugs. With automation, machines can perform in hours or days tasks that might have taken an individual months or years.
The Broads leaders argue that a new kind of organization is needed to enable researchers to access resources like these and answer the big questions that genomics poses. Traditionally, biology is organized into laboratories run by individual investigators and staffed with graduate students and postdoctoral fellows—young scientists in training who take on projects and then leave after a few years. Investing in new technologies is a major effort, so laboratories specialize in certain techniques, but not in others. Often the large-scale questions go unanswered, except by large centers that already have the resources to address them.
The Broad offers researchers at Harvard and MIT access to its technical capabilities, which are organized into platforms: facilities specializing in techniques like genome sequencing, genetic analysis, and chemical biology. Instead of relying solely on its graduate students and postdocs for technical skills, the Broad employs a dedicated staff of research scientists to run its platforms. These scientists, who have no academic duties, are wholly involved in collaborative research with other researchers.
Some scientists are concerned that centers like the Broad promote a bigger is better attitude that values massive data sets and expensive techniques over detailed investigation. Although the Broads studies are conducted on a scale beyond what a single lab can accomplish, its leaders stress that they do not push big science over traditional investigator-led research. David Altshuler says that, For most kinds of research, the form that follows the function is probably what its always been: the individual lab working with the professor, because the problem is best tackled in that way. The Broad, he said, is devoted to questions that are best answered by teams working together.
Lander says that the vision for the Broad came from unofficial collaborations between scientists at Harvard and MIT that emerged while he headed the Whitehead Institute, a major center for sequencing the human genome. As the project came to a close, this growing culture of collaboration was so powerful, we felt we needed a way to institutionalize it. The idea took years to become a reality. Eli and Edythe Broad, who were funding a minor project at the Whitehead, made a chance visit there on a Saturday and were impressed by the buzz of activity and conversations going on. After many discussions with leaders at Harvard, MIT, the Whitehead, and the teaching hospitals, they agreed to fund Landers vision in Boston with a $100-million founding gift, which was later doubled. (As this issue went to press, the Starr Foundation of New York City announced a $100-million gift to fund a five-year consortium linking the Broad to four New York research centers—Cold Spring Harbor Laboratory, Memorial Sloan-Kettering Cancer Center, Rockefeller University, and Weill Medical College of Cornell University—to further focus genomic technology on understanding and treating cancer.)
The Broad currently has an operating budget of about $125 million, with most of its funding coming from the National Institutes of Health (NIH) and other federal sources. It has won major federal grants, including $14 million to identify common genetic variations called single-nucleotide polymorphisms (SNPs); $18 million to study heart, lung, blood, and sleep disorders; and $12.6 million to study cancer genomics. About 5 percent of its budget comes from the two universities. The Broad has also sought partnerships with pharmaceutical companies for some of its projects. Whether part of a public or private partnership, results of the Broads work are all made publicly available.
Though MIT handles the administration for the institute, it is governed by both universities. The science community of Boston and Cambridge is already collaborative, and connections have been forged among schools and science departments at Harvard, between Harvard Medical School and its affiliated hospitals, and between Harvard and MIT. Some of these connections are informal, some are virtual research centers, and some are educational connections like the Health Sciences and Technology joint M.D.-Ph.D. program at Harvard Medical School and MIT.
The Broad aims to take collaboration a step further. Its founders say it is both a real and a virtual institution. Its sleek new building in Kendall Square provides labs and office space where people from different institutions can work together cheek by jowl, as Lander puts it. In addition to its full-time staff of 625 and six core faculty members (expected to reach 12 in the near future), the Broad has a network of more than 100 associate members in the Harvard and MIT community, helping it reach beyond its walls. Core members have faculty appointments and maintain full teaching responsibilities at one or both schools; Lander, for instance, has been teaching MITs introductory biology class for 15 years, and Stuart Schreiber, Loeb professor of chemistry, teaches organic chemistry at Harvard. They also supervise undergraduate and graduate students in their labs. New core members are hired through academic searches with university or hospital department leaders.
The Broad solicits nominations for associate members from department heads, school leaders, or the researchers themselves, and a group of 10 core and associate members act as a selection committee. Associate members keep their full teaching and research responsibilities, but shuttle between the Broad and their home institutions, sometimes maintaining labs at both places. Memberships have two-year renewable terms, and members are expected to collaborate with others, attend weekly program meetings, and otherwise contribute to the Broad community. In return, Lander says, they are given only the right to propose collaborative projects. With its private funds, the Broad maintains an internal grant-making mechanism that can help researchers launch projects without facing the onerous process of applying for grants from the National Institutes of Health. These internal funds can nurture early or risky projects until they can find support elsewhere.
Forging an institutional marriage between Harvard and MIT was challenging enough. But the Broad is also intended to help bring together Harvards widely separated science community. In July, a report from the University Planning Committee for Science and Engineering at Harvard noted how boundaries between institutions, schools, and departments inhibit collaboration (see Sweeping Change for Science, September-October, page 71). Lander calls the Broad a horizontal connector cutting through the vertical structure of isolated departments and institutions. Altshuler adds, We are not trying to create another silo in Boston, by which I mean an organization that exists for itself only. If the Broad becomes a silo, it will be a failure.
The Broads team-based structure allows young scientists to lead projects, and Lander believes this quality will determine its success. Certainly many of the Broads junior members have already benefited. Three junior members, with help from Lander and the Broad, were able to collect $18 million to build a resource for RNA interference, a new technology for turning genes off in cells. Normally, a senior scientist would be in charge of such a venture.
The collaborative approach is attractive to young scientists who like teamwork—though it challenges the traditional model of training in biology, which values well-rounded scientists who do much of their work independently. Pardis Sabeti, M.D. 06, for example, a postdoctoral fellow at the Broad who worked with Lander as an undergraduate at MIT and won a prestigious Burroughs Wellcome Fund Award after earning her medical degree, says that Lander has created this sort of can-do atmosphere, where everything is possible.
She explains that the Broad follows a collaborative, business-like model, in which people develop specialized skills and work together in teams. Sabeti, for instance, has developed techniques to identify areas of the genome that have been honed through natural selection. She also works with other scientists to analyze their data; for instance, she is part of a team led by Dyann Wirth, Strong professor of infectious diseases at Harvard School of Public Health and an associate Broad member, that studies how the genome of the parasite that causes malaria varies around the world. The team, which includes Harvard and MIT members, was able to reach an answer to its question in a relatively brief eight months of collective work.
Whats impressive is how the Broad draws together the human resources of Bostons scientific network, says Harold Varmus, president of Memorial Sloan-Kettering Cancer Center, former NIH director, and a member of the Broads scien tific advisory board. The institute is admittedly operating on a scale that most of us would envy, he adds, but there are other centers around the world doing similar work. But the Broad has become a magnetthat seems very powerful at the moment, he explains: collaboration between Harvard and MIT is particularly useful because they have complementary resourcesHarvard has clinical expertise and access to patient information, while MIT has strengths in basic sciences and computational science, which are needed to make sense of large data sets.
Lander is characteristically excited about the potential for such a connection. If you can harness the diversity of expertise by creating a community in which people are able to share and work together, then the Harvard and MIT community is just unstoppable, he says. It is just the most powerful research community in the world.