Daniel MacArthur at Genetic Future has an interesting post on one of 23andMe's research efforts--the genomics start-up is currently recruiting healthy athletes over the age of 50 at the Summer National Senior Games in Palo Alto. According to an article from Palo Alto online "the company wants to find the genetic factors for healthy aging." A tweet from 23andMe yesterday says that 4500 senior athletes have enrolled since March.

23andMe has been pushing its research agenda for the last few months, announcing a partnership with two Parkinson's non-profits in March, and launching the Research Revolution--a contest to win research efforts into a specific disease--earlier this summer. The company's founders spoke with me about their vision for an article I wrote last year about 23andMe:

[They] envision spurring a sort of grassroots research effort that mirrors the rising influence of patient-advocacy groups, such as those that have organized new research projects about autism and Parkinson's disease. If the trend attracts large enough numbers, people with particular diseases could come together to search their genomes for similarities. Or those who escaped a particular condition despite a high genetic risk could provide insight into lifestyle and other genetic factors that were protective.

MacArthur's post comments on the senior athlete project:

Now, it would be easy to portray this strategy in sinister tones - the evil corporation stealing the genetic secrets of elderly athletes - but it seems to me that 23andMe has been fairly open about their intentions here, and I'm also genuinely intrigued about the potential of this set of individuals as research subjects. Physical activity extending into later life is a powerful protective factor against a multitude of common diseases, and digging into the molecular basis of variation in late-life physical performance could provide some genuinely useful and health-relevant insights.

That's not to say that getting useful results out of this cohort will be easy; but it seems plausible to me that many of the interesting traits that could be mined for this cohort would be determined by common variants (for reasons involving natural selection and post-reproductive traits, which are a topic for another post), and if a reasonable fraction of the 4,500 recruits end up progressing into the research stage that's a decent-sized cohort to draw on.