Departing U.S. Genome Institute Director Takes Stock of Personalized Medicine

See allHide authors and affiliations

Science  06 Jun 2008:
Vol. 320, Issue 5881, pp. 1272
DOI: 10.1126/science.320.5881.1272

For 15 years, Francis Collins has been the face of the public effort to decipher the human genome. So it came as a surprise last week when the physician-geneticist announced he will step down as director of the U.S. National Human Genome Research Institute (NHGRI) on 1 August. He plans to write a book on personalized medicine and launch a broad job search.

Since taking the helm of NHGRI—then a center—in 2003, Collins led a then-controversial 10-year effort to sequence the 3 billion bases of the human genome, ushering in “big biology.” The Human Genome Project sparked a debate about whether genomic data should be freely available, an argument Collins championed and ultimately won. He has also been a tireless advocate of legislation to protect people from genetic discrimination. The passage of such a law last month was one “signal” that it was time to leave, he says.

The timing of Collins's departure has prompted speculation that he wants to advise one of the presidential campaigns, a possibility he says would interest him. Which side is not clear: He told Science only that he considers himself an independent. Collins is an evangelical Christian who wrote a 2006 book about his beliefs.

Collins reflected on the state of genomic medicine and his plans in a telephone interview last week. His remarks, edited for brevity, follow. A longer version, including his thoughts on dealing with the egos involved in the human genome program, is available at

Q: Genomewide association studies are turning up many new genetic risk variants for common diseases. Yet when people have their DNA scanned for these markers through companies like 23andMe, they learn that they have slightly elevated risks for diseases that they are at only modest risk of developing in the first place. If this is personalized medicine, it seems disappointing.

F.C.: I co-wrote a paper back in 1997 in Science basically arguing for the International HapMap Project [to map human genetic diversity] and for the potential that would have to reveal genetic risk factors in common disease. There were people saying HapMap was basically a welfare program for the genome centers. So the successes that you now see pouring out of the pages of journals are exhilarating. At latest count, I think 165 genetic variants associated with a common disease have been discovered in the last 2 years.

But let's be frank about it. The discoveries still only account for a fraction of the heritability of diabetes or heart disease. A lot of people have been speculating that variants that are less common—1% or 2% or 3%—will have a larger effect in people that are carrying them. So I think in another 6 or 8 years [when rare variants are known], the predictive power will ratchet up substantially.

I'm of two minds here. I'm actually quite excited to see all these discoveries that many people are interested in taking advantage of. But I'm also worried that the public might become disillusioned and say, “Is that all there is?” And that could set the field back.

Q: How will personalized medicine work 10 years from now?

F.C.: I think we'll be at the point where sequencing your genome or mine could be done in an hour for under $1000, and people will say, “No big deal; why did they ever think this was hard?” But we'll still have this question of, “Okay, we can now make a pretty good prediction about your risk. What should you do about it?”

And here is one of my major frustrations. We desperately need, in this country, a large-scale, prospective, population-based cohort study. And we need to enroll at a minimum half a million people. We would need to have their environmental exposures carefully monitored and recorded, their DNA information recorded, their electronic medical records included, and have them consented for all sorts of other follow-ups.

I tried very hard to get some enthusiasm for this at high levels in the Congress and the [Bush] Administration, but as soon as people realize that this is genome project-like in terms of its cost, they shake their heads and say, “Well, it's not the right time.” It would be $300 [million] or $400 million a year to run this study.

Q: Do you think the day will come when the large sequencing centers supported by NHGRI are no longer needed?

F.C.: Yes. They're incredibly valuable right now, but I do look forward to a time, maybe 5 years from now, maybe 7 years, where DNA sequencing is so exportable, and so much of a plug-and-play operation, that the need for these very large, complex centers fades away.

Q: Why are you resigning now and moving, as you put it, into “the white space of unemployment”?

F.C.: At age 58, I'm restless to see what other challenges might be out there waiting for me. I want some time to reflect on all that's happened, to step out of the pressures of a 90-hour work week, and see where is all this going. Maybe there's something I can do in global health or something that relates to the real applications of genomics to medicine—given my physician background, that's where my heart is.

This sort of systematic [job] search, where you talk to lots of people, is pretty untenable when you're an NIH institute director. That is especially true now with the very high scrutiny towards conflicts of interest.

View Abstract

Stay Connected to Science

Navigate This Article