Wednesday, August 16, 2006

Collins's wager   posted by JP @ 8/16/2006 02:02:00 PM

The transcript of Francis Collins's speech at the 2005 meeting of the American Society of Human Genetics was published in the most recent American Journal of Human Genetics. In it, he comments on the prospects of genome-wide association testing to find variants involved in complex disease. Not everyone thinks it will pay off as a stategy as much as Collins (as the leader of the primary funding agency for the HapMap) does. Here's his bet:
How successful will this strategy [whole-genome association] be? We don't really know. But I've recently made a wager with my former chairman, Tom Gelehrter. Tom is a professional cynic; I'm an incorrigible optimist. So, we're going to see who's right this time. I'm betting that, by the time of the ASHG meeting in 2008, whole-genome association studies will have led to the discovery of at least four validated-not just guessed at-susceptibility variants for at least five common polygenic diseases. I've asked Joel Hirschhorn to be the referee. And, of course, the wager is substantial: one beer of the winner's choosing.
The first time I (mis)read that, I was shocked that anyone would take him up on it. But on re-reading, it's clear he's saying there will be at least 20 validated susceptibility variants discovered in the next couple years-- four variants each for five diseases.

That's a hell of a wager--essentially he's betting that our knowledge of the genetic underpinnings of complex disease will take off exponentially in the next two years. But two years isn't such a long time--genome-wide association studies have only recently become feasible, so their results are still a little ways off. Still, though, I'm more of an optimist, I'd say I'm with Collins on this.

I might (if I remember and don't get bored with what is essentially stamp-collecting) try to keep score as the papers get published, and maybe someone will tell me if my final tab is in line with Hirschhorn's. The big key here is that the association studies have to be genome-wide, as opposed to in candidate genes. So far:
1. Obesity - INSIG1
2. Type I diabetes - IFIH1
3. Macular degeneration - CFH
4. Caridiovascular disease - NOS1AP (this study actually looked at an endophenotype, not the actual disease outcome, but I'm counting it anyways)