Dr. Eric Green, PhD, MD, is Director of the National Institutes of Health's National Human Genome Research Institute. The following is excerpted from an interview.
The pace of progress:
One thing I've heard said repeatedly about genomics in the 22 years I've been involved with it is that we tend to overpredict where we'll get in the short term, say three to five years, and we tend to underpredict where we'll get in longer intervals, like ten years. I think that phenomenon has been described by someone else in another field, but it really applies to genomics. It seems that over and over again, we are way overly optimistic about what's going to happen in three to five years, and yet every time we look back at what we've done in the last ten years, we're shocked by how far we've come. I think that's absolutely the case now, especially in terms of data generation and DNA sequencing technologies. There's no evidence I can see that it's going to slow down; I don't think genomics is going to hit the wall. I think it has as much momentum now as it did a decade ago, and I would contend that ten and twenty years from now, we will be even more surprised than we thought we would be. So I guess one of my overarching comments is that I see no reason to think that the pace at which we are developing new technologies, understanding our genome, and figuring out how it's going to be medically relevant, will slow down.
Genome sequencing and analysis:
One thing I would predict is that the technologies for generating data will create a situation where data generation is trivial and analyzing the data becomes the overwhelming challenge. I think genomics is going to become more and more an information science and less a technology science, and I think the great challenges are going to be in how we analyze and interpret data in creative and powerful ways; and every time we need to generate more data, that will be the least expensive part of the equation.
Once upon a time, the Human Genome Project was all about data generation; now we already find ourselves in a situation where we have data abundance but an analysis restriction. That disparity between the amount of effort to generate data and the amount of effort to analyze it will only grow with time. I can imagine 20 years from now it might cost $500 or $100 to generate a genome sequence, but to fully interpret it might cost more than the sequencing costs.
Discovering how DNA works:
The second prediction I have-a very bold prediction-is that 20 years from now, we will still be discovering basic ways that DNA confers function. I do not believe 20 years from now we will have figured out every last way that DNA encodes biological information; I still think there are major surprises out there to be found. I think there are major mechanisms still to be discovered, and with that will be a continued need for strategic interpretation.
I think there's a lot of biological information encoded in DNA that we will still be discovering. I'm even saying that we're going to be discovering basic mechanisms in 10 or 20 years. Even if we say that we think we know all the promoters in the genome, I'm sure 20 years from now, we'll still be discovering new promoters acting in ways that we didn't know about.
I always say that the human genome sequence is like a great novel. We'll be spending dozens and dozens, maybe hundreds of years interpreting and re-interpreting it, just like a great historic novel. It's naïve to think that even in 10 years or 20 years we'll have a complete catalog of every functional sequence and any deep understanding of how it works.
A revolution in evolutionary biology:
My third prediction under the general research area is that we will see a completely new way of studying evolutionary biology that will be fully computational. I think 20 years from now, probably before then, we will have genome sequences of thousands and thousands of animal species. A 10th grade biology student's laboratory exercise will not be confined to dissecting frogs or looking at a fossil; they'll be sitting at a computer and will have tools in front of them to look at genome sequences of tens of thousands of different vertebrates, and their laboratory exercise will be to figure out how DNA changes have led to biological innovation.
There will be almost an entirely new field, a subcomponent of evolutionary biology, that will be dominated by computational analyses. Yes, we'll still be digging up fossils, we'll still be doing imaging and biometrics, but we will also have in front of us a database of tens of thousands of genome sequences from all different kinds of critters that walk and swim and fly on this earth. Just imagine the experiment where you can look at a given stretch of a genome and trace the evolutionary history of every little piece through tens of thousands of vertebrate genomes. It's incredible, but it's absolutely doable 20 years from now.
Genomics in medicine:
I believe that certainly 20 years from now, the use of genomic information about individual patients will be standard of care. I think when it comes to cancer, it will be pervasive; I think genomic-based analyses of cancers will become standard of care for many different kinds of cancer probably well before 20 years. For pharmacogenomics, it will be standard of care for dozens, if not hundreds, of different conditions for which we will use genomic information on patients as a guide for selecting and dosing medications. And I'm very confident that we will use genome sequencing as standard of care for diagnosing rare single-gene genetic diseases.
Hand in hand with that, I can believe that the routine will be that you'll have a genetic sequence of every patient. Now, we can start wondering what it will look like, whether that genome sequence is obtained as part of newborn screening shortly after birth ... I realize there are still many complicated issues, but I think one can certainly envision that whole-genome sequences might be generated as part of newborn screening.
I can't believe that electronic health records won't be standard of care in hopefully most places in the world; and I can't believe that genomic information wouldn't just flow into those electronic records. But, again, that is another area where there are lots of complexities and questions, and we're doing research in that area to clarify things.
Where I'm less certain is what the role of genomic information will be for truly understanding the genetic basis of common complex diseases in terms of individual patients. I don't know whether we'll get to the point in 20 years where we can look at 100 different loci and say, 'You are at a 42% greater likelihood of getting coronary artery disease, and this is what you should do.' I think the jury is still out on what that's going to look like, and I wouldn't want to overstate that part. I think that's going to be a question mark for now.
Understanding gene-environment interactions:
I believe we will gain a much more sophisticated view and understanding of gene-environment interactions. On the genomics side of that equation, the technology surge has really happened in the last decade and will probably continue over the next decade; but I think that we're getting to the point where it's going to become trivial to gather data about the genome. I think the surge to anticipate over the next 10 or 20 years will be technologies for doing environmental monitoring. I think that one of the reasons we're ignorant in understanding the environmental basis of disease is that we just don't have technologies for doing fine-scale measurements of environmental exposures. I'm not sure my field is going to have anything to do with it-I don't think it's genomics, I think it's environmental science-but I think technologies are coming; and with that would come much more powerful studies to capture data on the environmental side that's just as powerful as the data we're getting on the genomics side.
Ethical, legal and social issues:
Finally, I firmly believe that the societal issues that we are already starting to grapple with around genomics-the ethical, legal, and social issues-will continue to require significant attention, significant research, and significant debate. I don't think these ethical issues are going to go away. With technological advances and increasing knowledge will come a continued need to wrestle with very hard questions. I don't think the questions are going to become simpler; if anything, I think they will become more complex. We shouldn't fool ourselves into thinking that we'll eventually figure all this stuff out and the ethical dilemmas will go away. I just don't think that's true. It's not that I'm pessimistic; I think we can deal with them-we just can't ignore them.