By CRG staff - interview with Dr. Eric Green, Director of the National Human Genome Research Institute
Eric Green, PhD, M.D., was recently appointed Director of the National Institutes of Health's National Human Genome Research Institute.

I understand you worked on the Human Genome Project since its beginnings - how did you start out?

I was a postdoctoral research fellow in Maynard Olson's laboratory at Washington University at St. Louis when the Human Genome Project started. I got involved in Washington University's Human Genome Center, one of the initial genome centers that were established for the project. In fact, I even presented some of the proposed work during the site review of the grant that funded and created the center. I felt like I was literally at the starting line. I didn't have a full appreciation for the historic nature of it at the time, though.

Dr. GreenHave you found at the National Human Genome Research Institute (NHGRI) that there is a push to advance 'practical' versus 'theoretical' research?

I don't necessarily divide research into theoretical versus practical. In some ways, it's more the case of basic versus clinical research. The Human Genome Project was very focused on getting the sequence of the human and other less complex model organism genomes, such as the roundworm and mouse, to give us insights into how genomes work in general. If you read NHGRI's April 2003 publication in Nature, "A Vision for the Future of Genomics Research," you'll find that our major focus following the Human Genome Project was mostly to address basic research questions.

Right now, the genome research community continues to try to understand how the genome functions at a basic level. That basic understanding will help researchers make connections with genetic disease research and clinical problems. As we look towards the future, opportunities to translate that information into clinically relevant applications are increasingly going to become the focus of genomics research. NHGRI has helped to develop a broad base of knowledge, tools, and technologies that we can use to start tiptoeing into clinical research arenas. We need to better understand how to integrate genomic information into studies of disease and, ultimately, into efforts to improve peoples' health. That's a huge challenge.

So 'translational' research means not only translating knowledge of the human genome into diagnosis, but taking the next step into therapy … should I use the word 'therapy'?

Yes, translation encompasses all aspects of clinical care, including therapy. But we're not yet at the point where genomics is a routine part of clinical care. We have a long journey ahead of us, with many paths still to be explored by both basic and clinical researchers.

At NHGRI, we've been thinking about translation for a long time because so much of what we do in our intramural program is to find ways to apply genomics to clinical research activities. That kind of translational activity is often first done at a place like NIH because we have an infrastructure for doing clinical research that is second to none. The NIH Clinical Center is the world's largest research hospital. Every patient in that Center is on a clinical protocol.

How do you decide when an idea is 'ready for translation' - is there a formal way of deciding when it's appropriate to bring human subjects into it?

It's not really a formal thing - you take a fundamental research approach to it.

I'll give you an example. How do we understand what it is going to be like when a patient is provided with his or her genome sequence? That's a fundamental question, right?

Of course, there are already companies that are providing genetic testing and information directly to consumers. A number of years ago, we were wondering if sufficient research had been done to provide insights into how physicians would use such genomic data or how patients would respond to being presented with their genomic information. We found that more research was needed.

So, we started the ClinSeq project at the NIH Clinical Center, where we're enrolling 1,000 individuals initially. We are studying what we learn from their genome sequence data, how that information might guide clinical care, how the participants deal with genomic information about themselves, and how physicians manage the information.

ClinSeq is about capturing information and data about genomic medicine in a research context. One of our goals is to talk to the participants before they enroll and to study how they respond to the notion of getting genomic information about themselves. We want to hear what they think they're going to do with the information when they get it, and in some cases actually give them some of the information. Meanwhile, we will eventually sequence their entire genomes and learn how to deal with such data for addressing clinical research questions. We've enrolled almost 800 participants to date.

That's the kind of framework that is needed to study the questions that everyone has about genomic medicine and get some answers before genomic sequencing becomes standard practice in healthcare.

And it does seem that direct-to-consumer genetic companies can circumvent some of that...

Well, it's not really that the companies are circumventing it. They're not doing anything illegal; they have technologies in place that can generate and provide genomic information to individuals. The real concerns revolve around what individuals, the medical system and society will do with this genomic information after they get it.

So, the ClinSeq study is both about the actual data gained from sequencing these volunteers' genomes as well as a sort of psychological study of their reactions.

Right, part of our study involves analyzing information about peoples' phenotypes and their genome sequence data. However, another aspect is that we're interacting with these clinical subjects around the notion of gaining genomic information and capturing those interactions. We're in a position to study them long term. And ClinSeq is just one example of the kinds of things we're going to need to continue doing in the field of genomics.

Another centers on genome-wide association studies- efforts to find regions of the genome that harbor variants conferring risk to complex diseases. There are some variants that tell you whether you are at higher risk for this or that disease, and coming out of that are opportunities for companies to provide that information to patients. In some ways this is exciting, but it's an oversimplification to think that you can just provide information and that's all. Each genetic variant confers a small amount of risk that needs to be analyzed within the context of many other things, including someone's health, family history, environmental exposures, and so forth. We need research to help figure out the best ways to advise people and to help medical professionals deal with this new frontier.

This is what NHGRI has been all about, and what it will be increasingly doing in the future. It's very different from 20 years ago, when the institute had one clear goal: to sequence the human genome.

So this is both the vision that you have and the direction that the Institute was already moving?

Absolutely. That's not to say that we're going to stop doing what we're already doing, because believe me, we don't completely understand how the genome works. In some ways, we're barely scratching the surface. The good news is that we are getting better and better technologies to help us do this, so what we could have learned in two or three years with the technology we had five years ago now can be learned in two or three weeks.

How central is epigenetics in that research?

It's one of many important sub-areas. Once upon a time, we thought all the action was in the genes and that was where all the disease-relevant variants were going to be. But then we realized there's also a lot of other, non-coding DNA that doesn't code for proteins, but is involved in a whole host of other things. Then, there's also a lot of variation not just in single nucleotides, but even in the number of copies of a stretch of DNA. All of a sudden, if you have four or five copies of some stretch of DNA, you have an increased risk for a disease. But that's not all of it: that doesn't account for all of the epigenetic marks that are put down on our DNA. And believe me, as of today, we do not understand all of the ways that DNA can confer function - it's crazy to believe otherwise!

For example, just a few years ago, we hardly knew much about microRNAs, and now it's the rage. We now know it's not as simple as DNA makes RNA makes protein. Many RNAs are made that do things other than code for proteins. And I think there are many other surprises in the genome if we keep searching.

There have been numerous studies claiming to find 'the gene for' this or that. I'm wondering, as these new complexities come up, if that research will be revisited or reconsidered.

Oh, absolutely - and the environment is a whole other component here, too. Our view of the genetic architecture of disease is undergoing a major metamorphosis. It's more complicated than we ever appreciated.

But it's compelling. This is not some geeky basic science research that people don't identify with. It's about understanding the complexities of disease. People will be very interested to see us work this through, even if it's complicated and takes a long time, because it's clinically important.

Is there much research happening at NIH and NHGRI looking at ways in which genes may be taking more of a backseat to environmental factors than previously thought?

Definitely, there are various studies, and probably more to come, where we are going to integrate environmental and genetic information. In fact, the current NIH Director, Francis Collins, has published papers proposing large population-based cohort studies. These are very expensive studies, though, because along with collecting genetic and health information, they involve collecting information about environmental exposures. To do this right, such a study needs very large numbers of participants and it needs to be prospective.

In the press release, you mention that part of your job is to 'push the application of genomics into all areas of biomedical research.' Correct me if I'm wrong, but it almost sounds like part of your job is sort of selling the technology?

I didn't mean to give that impression, perhaps 'facilitating the dissemination of genomics' into other research areas would have been a clearer way to put it.

Let me give you an example. Did you realize there were more Recovery Act funds given to our big three sequencing centers by other NIH institutes than by NHGRI? That's sort of cool, right? It means that other NIH institutes working on specific diseases - whether it's heart disease or mental illness or diabetes - are looking for the most compelling opportunities that will help advance their research agenda. And they voted with their feet. They infused significant amounts of money into the three sequencing centers that NHGRI has been developing and supporting for the last 20 years. It shows we have created knowledge and expertise that's valuable not only for understanding the genome, but for studying complex diseases like cancer and autism. We regard that as a complete success.

You also mention collaborating with a wide range of researchers, in 'all areas of biomedical research.' That's a very large space - do you really mean 'all areas'?

Well, I can't think of an institute at NIH that isn't somehow using genomics. Actually, I'd go beyond the NIH - think of the National Science Foundation or the United States Department of Agriculture. Genomics is everywhere, whether you're studying disease or evolution or how to make better crops. Genomics is a very integrative discipline.

Is NHGRI working much with behavioral medicine?

Without question, this is part of what we do. One of the first things I did here about six years ago as the NHGRI Scientific Director was to create the Social Behavioral Research Branch. The social implications research now performed by this branch involves addressing some behavioral questions relating to genetics and genomic research. Such research is now starting to be performed by other institutes as well.

You were talking earlier about how much more complex connections are between genomics and phenotype than we thought, and it seems that social and behavioral manifestations of genes would be especially complex.

Certainly. The scientist who heads our Social and Behavioral Research Branch, Colleen McBride comes to mind - her previous area of research dealt with smoking cessation. The common theme is that just because you give people information, it doesn't automatically lead them to change their behavior in a healthy way. People know that smoking is not safe, so why do they do it? That might not be very different from someone in the future getting definitive information that they have a genetic predisposition to a disease. How will that information change their behavior? Research in this area is going to be critically important for implementing genomic medicine.

Considering how broadly genomics can be applied - and considering how big your budget is this year - how much does NHGRI seek out projects to fund rather than waiting for the projects to come to you?

Trust me, we have far more people who want to interact with us or get grant money from us than we can accommodate. That's not a problem. The other part of the mix is when we come up with ideas and put out a request for applications in that area. Like most NIH institutes, it works both ways, but we are an institute that has a reputation for developing large multi-investigator initiatives that we want to pursue.

The Human Genome Project was a classic example. NHGRI was initially created to lead that project. After that, we launched a whole series of initiatives that built upon the foundation laid by the human genome sequence. For example, we started an initiative called the Encylopedia of DNA Elements (ENCODE) to catalog and understand the functional elements of the human genome. We also funded research to develop innovative technologies to sequence a human genome for $1,000 or less; the HapMap Project, which developed a map of common human genetic variation; and the Genetic Association Information Network, a set of genome-wide association studies to help us understand the genetic basis of complex diseases. We tend to be more initiative-oriented - but if a researcher comes to us with a great idea, we look to see how we can make it happe n.

Share this page:
Facebook Twitter StumbleUpon

Search: GeneWatch
The Gene Myths series features incisive, succinct articles by leading scientists disputing the exaggerations and misrepresentations of the power of genes.
View Project
Other Genetic Issues
View Project