I met Cynthia in a van from the airport, headed to the annual meeting of Family Tree DNA (familytreedna.com), where I was to speak about genetic testing. A beautiful blonde who looked decades younger than her 60 years, she’d led a painful life, with type 1 diabetes since childhood, just like her father, brother, and grandfather. The family, so they thought, was 100% European, mostly Polish.
My talk did not go over well. Genetic testing companies and their customers do not like to hear that a geneticist thinks their tests should be regulated, for reasons of both privacy and accuracy.
Cynthia, intrigued despite my warnings, sent off a spit sample to 23andme (23andme.com), to learn about her ancestry. She got that, and more – health information, including a “lower than average” risk of developing diabetes. deCODE Genetics (http://www.decodeme.com/) gave her the same answer. Ditto her brother.
But her brother’s Y chromosome held an explanation. About 1200 years ago, a Korean man and at least two Chinese men dropped a bit of DNA into the family. So when Cynthia went back to 23andme and recalculated, entering “Asian” instead of “European,” her diabetes risk shot up to 90%.
So it looks like ancestry testing helped get this family on the right track. But another way to look at it is that the health-related tests are simply not precise enough.
This past week “direct-to-consumer” genetic testing took a hit, and it’s about time. First the Walgreen’s near-fiasco of off-the-shelf direct-to-consumer genetic tests, then a white paper from the American Society of Human Genetics calling for oversight of ancestry testing. To top that off, I got a call from a writer for a popular psychology magazine asking me for a “sound bite.” A sound bite? Genetic testing isn’t quite that simple.
Genetic tests for well-studied mutations, delivered by a genetic counselor or physician, in person, are fine. But the genetic “associations” gleaned from population data, although useful in research, often cannot reveal much of anything about an individual – such as Cynthia.
[cross-posted from Ricki Lewis' blog, GeneticsWatch.]