23andMe and moderate risks
At Think Gene, Andrew Yates has a review of 23andMe. Mainly about the web interface — that does look cool — but then he concludes:
The Disappointing
Of all the 70 possible traits described by the 23andMe test, no trait reported anything useful except perhaps a moderate (almost 30%) risk of rheumatoid arthritis.
The sad fact is that, for common SNPs, all risks we know of are small to moderate. If we knew how they interacted, maybe we could detect higher risks, but right now we only see small increases in risk in the SNPs we know about.
If you expect more, you will be disappointed indeed.
I think personal genomics is pretty cool, but I doubt it has any clinical relevance at this point. I could be wrong, of couse, but that is where I would place my bet.
June 8th, 2008 at 7:44 am
Yes, I think that 23andMe is more like a placeholder for an invaluable product that will exist in ten years… much like the original personal computers were largely useless, expensive toys, as were websites.
June 8th, 2008 at 8:26 am
Oh in ten years I am optimistic that we will know a lot more about genetic disease risks and probably be well on our way to personalised medicine.
June 8th, 2008 at 11:37 pm
That’s why we have to build brick and mortar practices which can sustain themselves first!
-Steve
http://www.helixhealth.org
June 9th, 2008 at 5:34 am
But, Steven, how much use do you see this genomic information — at present — to be in a clinical setting. Enough to try to set up a business on it, I get that, but concretely, how will genetic information help me, as a patient?
The genetic factors are tiny compared to environmental factors (for the variants we know about, anyway). Will people change their habits if they learn that they are in a high risk group because of their genes, if “high risk” is a relative risk of 1.2?
I see the point of personalised medicine, but I don’t know how far we are from being able to pick medicine based on genotypes…