I mostly agree, although 23andMe did tell me something very useful. I have the two G alleles for rs4680. (http://www.snpedia.com/index.php/Rs4680) One research paper said I am likely to respond well to Modafinil. I now find Modafinil extremely useful. While 23andMe is mostly for the coolness factor right now, it's only going to get better. For a small fraction of people, data from services like 23andMe is undeniably useful, since it will tell them things we do know with high probability (drug interaction/metabolism, early-onset Alzheimer's from ApoE4, etc).
The best probability estimate for most adult diseases still comes from history and the physical exam, not from genetic tests.
I meant that genetic testing along with a physical exam is more useful than just a physical exam.
We are getting closer to clinical utility of these tests. I'm no Luddite. But I think that these tests need to be thought about like every other medical test. (See http://www.nejm.org/doi/full/10.1056/NEJMoa0706728 for why I say that we're not quite there yet, but getting closer for some phenotypes.)
It's one thing to say that you have a right to your raw genetic data; like I said, I'm sympathetic to it from a rights-based standpoint, though not a clinical one for reasons that bhickey has brought up. It's another thing to say (not implying that you are saying this) that companies should be allowed to make clinical claims about you based this data. That interpretive step merits thoughtful consideration, and probably regulation.
Getting GWAS data from 23andMe, on the other hand, is generally not so much useful as it is entertaining.
The best probability estimate for most adult diseases still comes from history and the physical exam, not from genetic tests.
> Also, there's the issue of rights. It's my DNA. I have trillions of copies of it in my body. I am literally made of this stuff. If human beings have any right over their own bodies then they should be able to read their DNA.
I am most sympathetic to that argument.