Newsflash: Rivers Do Not, In Fact, Run Uphill

[nyuanshin]

This article heralding the death of the "common diseases caused by common genetic variants" hypothesis provides me with yet another cause for nervosfreude. It's a little incredible that it took people this long to figure out that common diseases are unlikely to have etiologies that are primarily genetic: as John points out, the few exceptional

Comments

an old email

[gcochran]

9/21/2002 2:36 PM

To the extent that genes cause a Darwinian illness, one
that reduces fitness, we expect that the fitness load of any individual
allele will not be very big. That means that we do not expect to find the
sort of allele that would be interesting enough to justify the search: we do not expect to see a common polymorphism with a strong effect that results in a Darwinian illness. I mean, how's it supposed to stay common? The common-disease-common variant notion is _never_ going to work except when the syndrome's effect on fitness in ancestral environments was
insignificantly small. That might be the case for a common problem that
strikes in extreme old age - like Alzheimer's - or for some syndrome that is moderately common and today and reduces fitness, but that is triggered by something new in the environment, and was insignificant in ancestral
environments. In that second case, of course the thing to do is to identify
and control the new environmental insult.

Gregory Cochran

Re: an old email

[nyuanshin]

Yeah, I know you've probably been calling this one for ten years. Do you know if anybody else with any actual clout was arguing the same thing? It's hard to understand how *nobody* in the field could have been standing up and saying this made no sense all that time.