Companies that offer saliva-swab DNA tests to assess your ancestry and health are rising in popularity. But do they really work?
3. While it can be obvious when DNA samples are not a match, definitively linking your DNA to that of your ancestors through similarities is not always so black and white. Assuming with each descendent, half of someone’s DNA is passed on to the next generation, you’re unique contribution is already down to below 1% after only ~7 generations. If I want to know if I’m related to Galileo, it will thus be challenging to link specific aspects of our respective DNA samples that aren’t also shared with just about everyone else. We do have mitochondrial DNA which is passed on only by our mothers and DNA linked to Y-chromosones which is passed on only by our fathers. These types of DNA can serve as more direct links to distant ancestors, but those ancestors only make up a small percentage of our lineage.
4. Interpreting your DNA ancestry results requires understanding confidence intervals. Most companies lay out your results in terms of how confidently you can trust them in terms of probability. Some even allow you to change those confidence interval settings to see the range in your results. For example, my results may tell me that I am 2% Japanese at a 75% confidence interval but that probability of Japanese lineage may increase if I’m willing to look at less confident results. These confidence intervals also mean that any results presented as accurate to the decimal point are probably less certain than they appear to be.
5. Understanding the limitations of percentages and probabilities is especially important in the interpretation of any findings-related disease mutations linked to genetic diseases. For example, having a particular gene variant known as N370S means you are three times more likely to develop Parkinson’s, which can sound intimidating. But the normal risk is only around 0.3% so a tripling of that risk brings you to still less than 1%. Not to mention much of the risk in developing Alzheimer’s or Parkinson’s is thought to be nongenetic or from genes that aren’t tested by companies like 23andme. Also, consider how it will affect you to know that you have a 5% chance of colorectal cancer, for example. 5% is still fairly low but worrying about such a result could be a source of stress.
Results require generalizations about genetics of large populations of people throughout time and across the globe.
6. Testing your DNA for disease markers should never replace seeking or continuing any form of treatment. The most thorough analysis of your genetics is best done through a face-to-face appointment with a genetic counselor who can personalize (and thus improve the accuracy of) your results by, for example, incorporating your family history of disease.
Do ancestry DNA tests work?
The answer is both yes and no. Direct-to-consumer DNA tests typically work as promised in the fine print—they offer a look at which regions of the world are home to people with DNA signatures most similar to yours. For many, that is enough.
However, those results require generalizations about genetics of large populations of people throughout time and across the globe. Will they be able to tell you that you are for sure related to Cleopatra? No. For those answers you’ll have more success digging through your family history. Our ethnicity is also not entirely dictated by our genetics. We each have a collection of lived experiences and cultural ties that also help play an important role in who we are.
Until next time, this is Sabrina Stierwalt with Everyday Einstein’s Quick and Dirty Tips for helping you make sense of science. You can become a fan of Everyday Einstein on Facebook or follow me on Twitter, where I’m @QDTeinstein. If you have a question that you’d like to see on a future episode, send me an email at firstname.lastname@example.org.
Image of double helix © Shutterstock.