December 1, 2013 (Vol. 33, No. 21)
Ten years after its “official” completion, what can be said about the Human Genome Project (HGP)? While the question is still debated in a few quarters, most of the bioscience research community and much of the medical community regard the HGP as a foundational achievement.
Not only did the HGP achieve its stated goal—sequencing the human genome—it has also had consequences that continue to ripple outward, changing the way we think about biology, changing the way we pursue medicine, and unleashing a host of technological and commercial initiatives.
By establishing the order of A’s, C’s, G’s, and T’s in the human genome, the HGP essentially set the stage for functional genomics research and clinical translation. Clearly, the HGP has far-reaching implications. And, just as clearly, any assessment of the HGP’s legacy should do more than review the past. It should look forward, too. Such is the case with this assessment, in which GEN summarizes the views of five distinguished scientists:
Mark Adams, Ph.D., Scientific Director, J. Craig Venter Institute.
David Botstein, Ph.D., Director of the Lewis-Sigler Institute for Integrative Genomics, Princeton University.
William Evans, M.D., CEO, St. Jude Children’s Research Hospital.
Geoffrey Ginsburg, M.D., Ph.D., Director of Genomic Medicine, Duke Institute for Genome Sciences and Policy.
Eric Green, M.D., Ph.D., Director, National Human Genome Research Institute (NHGRI).
A decade, of course, is too short a time for anyone to attempt a definitive statement of the HGP’s place in history. Nonetheless, it should be possible to form a general impression, even in an article as brief as this one.
Other reasonably brief but informative resources include: a compilation of HGP articles on the NHGRI’s website (see “All about the Human Genome Project”); a review of translational medicine accomplishments and prospects (see Dr. Ginsberg’s optimistic viewpoint published in JAMA April 10); a warning that the kinds of basic research supported by the HGP could be deemphasized in favor of applied research (see Dr. Botstein’s November 1, 2012 article in Molecular Biology of the Cell).
GEN asked four questions of the panel: Why should completing the human genome project be considered such a historical accomplishment? Did the completion of this project meet expectations? What were the main surprises that have come from studying the human genome? What next needs to be done to advance our knowledge and to capitalize on our understanding of the human genome?
The HGP’s Place in History
Today, the human genome, genomics (a rarely used word 15 years ago), and sequencing inform virtually all of biology. GEN’s panelists agree the HGP was transformative, though perhaps differing on particulars. Much of what followed, they say, is derivative—intended to better exploit and explain how genetic code works.
The HGP was biology’s version of the U.S. moon shot, says Dr. Botstein: “You could argue there was no actual practical use for going to the moon; however, the technology that was built to take us to the moon changed society in many ways. I would say that the HGP paralleled this achievement, except that with the HGP, it was clear from the beginning that the sequence would be valuable.”
Insight into human biology flowed almost instantly from the genome; translation into medical practice has taken longer. Still, “ten years is a blink of the eye in terms of medical research. I think it’s been quite impressive,” offers the NHGRI’s Dr. Green. A trickle of genomics-based drugs, mostly in cancer, has reached the market—for example, the breast cancer drug trastuzumab (Herceptin), which only works for women whose tumors are HER2-positive, and gefitinib (Iressa) and erlotinib (Tarceva), which help patients whose tumors are positive for EGFR mutations.
Moreover, declining sequencing costs are accelerating the search for new drugs and pushing sequencing into the clinic. In 2010, St. Jude’s Research Hospital started the Pediatric Cancer Genome Project, a three-year project to sequence normal and cancer cell genomes of 600 pediatric cancer patients. It reached this goal ahead of schedule, having completed sequencing for 700 patients (accumulating 1,400 whole genomes) to date.
“We couldn’t have done this without the HGP,” says Dr. Evans. “We’re using next-generation sequencing, and our reads are 100–150 bp. If we didn’t have a nice framework to align these genomes on, this work would be close to impossible.”
“I see the sequence and sequencing tools as foundational. When we started, we were anticipating the average cost of each genome over a three-year period would be $35,000. That’s probably about what it turned out to be. Now we’re well under $5,000 per genome and rapidly heading toward $1,000 per genome,” adds Dr. Evans.
A follow-on project will allow St. Jude’s to sequence patients’ cancers in real-time and use the information to inform therapy decisions.
It’s also worth noting that until the HGP, biology was a classic “small science”—dominated by individual investigators or small multi-investigator group working in relative isolation. That’s changed dramatically. Given the cost and scope of the HGP, collaboration was necessary. Today there are many consortia-based efforts—the International HapMap Project and the 1000 Genomes Project, to name just two—pooling data and coordinating research.
The HGP has changed the discipline in yet another way: once primarily descriptive, biology has become a Big Data science. There’s no going back. New IT and bioinformatics tools are constantly in development, and they have forced researchers to retool skill sets. Moreover, the ongoing emergence of less-expensive, desktop-scale sequencers will give smaller labs a seat at the Big Science table. Ultimately, what it means to be a biologist and what tools one must be competent to use were forever changed by the HGP.
Says Dr. Botstein bluntly, “I think that the HGP exceeded every reasonable expectation for those who understood the issues and those who actually did the heavy lifting. I think that in the middle of this process, because of concerns about funding, people began to escalate expectations beyond reason, and of course those unreasonable expectations can never be met. So I think all the hoopla—the race between the public effort and J. Craig Venter’s private effort, the White House meetings, and all the rest—is where people kept raising the bar beyond what was ever realistic.”
It is probably sufficient to say, yes there was hype as well as a simplistic view by many regarding how quickly the HGP would pay off—whether in healthcare, profitable enterprise, or even scientific insight. “The 90s were a time when there was a lot of hype around the genome project, frequently driven by business aspirations and a few scientists as well. That produced maybe some false expectations on the part of the lay public and maybe some disappointment at the end,” says Dr. Ginsburg.
There have been a number of unexpected findings, starting with the number of genes, which at roughly 20,000 is far fewer than the 50,000–100,000 first predicted. Dr. Botstein says wryly, “I had some bets, which I lost. I, like most people, was too high. But nobody got the right number. However, it has to be said that in retrospect there are simple ways to have calculated the right number.” It does seem that organism complexity has little to do with genome size.
Another critically important scientific surprise, Dr. Botstein says, “is the detailed way in which we can follow the evolutions all the organisms in the world back to their common ancestors. The detailed process of evolution is easily and quantitatively made visible by sequencing. I think we all sort of understood that such results were there, but when you actually look at the tree of life and all the connections and sequence similarities, it’s truly astonishing. You can’t think of evolution as a theory anymore.”
Here are three other notable surprises:
- Not so much junk. While just 1.5% of the human genome encodes proteins, comparative analysis shows another 5–8% is highly conserved over the millennia. “We don’t exactly know what it does. It makes RNA, and people are just beginning to work out what all that RNA transcription means,” says Dr. Adams of JCVI. “The harder we look, the more functional elements we find.” Indeed, the controversial Encyclopedia of DNA Elements (ENCODE) Consortium suggests that the number of functional elements in the genome may be far greater than once thought.
- Microbial company. It turns out that the human genome is surrounded by thousands of microbial genomes. In fact, in any given individual, the human genome is probably dwarfed by those genomes in terms of numbers of nucleic acids and species. Human cells are outnumbered 10 to 1 by microbial cells. “This is a really surprising and fascinating finding, and we are just beginning to understand it. We are sort of an ecosystem of genomes as opposed to a single human genome,” says Dr. Ginsburg.
- Genome-wide association studies. “When GWAS started back in 2005,” notes Dr. Adams, “there was every expectation they would identify powerful risk markers for disease in the clinic, and we found very little compared to what we expected.”
Dr. Evans adds, “Everybody is now realizing that we probably found a lot of the low-hanging fruit. Going forward, it is going to be harder. We are going to have to increasingly focus on the rarer variants as well as the common variants that are collectively influencing disease risk or response. You know the analogy of gold dust versus gold nugget: If you put enough gold dust together, you have as much value as one big gold nugget.”
Sequencing technology has raced forward, but further advance is needed. “Bisulfite whole-genome sequencing is the strategy we will use going forward, but it’s far from perfect in terms of identifying methylation sites and epigenetic changes,” says Dr. Evans.
Dr. Adams agrees, “I do think there is still technology development to be done on the sequencing side to get long reads. Will it be done by Oxford Nanopore? Will it be done by Pacific Biosciences? Who knows? There is still a black hole—we don’t understand structural variation in individuals and among cells.”
Data management and analysis remain thorny issues. Moreover, ensuring ready access to data for researchers and clinicians is critical. And, while the new crop of smaller sequencers will enable individual investigators, Dr Ginsburg worries interesting findings may end up locked in someone’s e-lab notebook. Better clinical decision support tools connected to electronic medical records are also needed, says Dr. Evans: “There’s no way physicians can remember all of this. There’s no way anyone can.”