What DNA ancestry tests can — and can’t — tell you

Video Statistics and Information

Video
Captions Word Cloud
Reddit Comments
Captions
So I recently took one of those at-home DNA ancestry tests. All I had to do was fill up a vial with a disgusting amount of spit and mail it off for analysis. We're gonna be here for a very long time. I just spit it back up in my nose. A couple weeks later, this is what I got: It's a neat little pie chart with these specific percentages that were color-matched to different regions on a world map. The report told me I was mostly Southwest Asian — no surprises there, considering both my parents are from Iran. That percentage — 86.7% — I understood that to be the portion of my DNA that’s West Asian. But it turns out, that’s not exactly what ancestry tests are telling us at all. This is an ad for one DNA ancestry test, 23 and Me. An ethnically ambiguous woman travels the world, and a circle animates around her, sort of like the pie chart in my test results, as if to say, this woman's DNA is 29% East Asian. And here's an ad for a different ancestry test. “52% of my DNA comes from Scotland and Ireland.” And somehow this information compels him to... wear a kilt? Alright, so what are ancestry tests really telling us? Can you help me understand what my results are telling me? Because I’m getting mixed messages from ads and how other people talk about their results. This is Wendy Roth. I'm an associate professor of Sociology at the University of British Columbia. OK. First of all, these test results are not about your entire DNA. They're about a tiny, tiny fraction of your DNA. To understand how genetic ancestry tests work, let’s start with the DNA itself. There are about 3 billion base pairs in our genetic code. Those are the As, Cs, Ts, and Gs that form the instructions that make us… us. Of these 3 billion base pairs, 99.9% are exactly the same in all humans. But for the remaining .1%, one person might have an adenine where another person has a guanine. These single-letter differences are called Single-Nucleotide Polymorphisms or SNPs. Groups of SNPs can help explain why some people are taller than others or why some people have green eyes while others have brown eyes. But most SNPs have no known effect at all. What many DNA tests are looking at are a relatively small number of SNPs, specific positions in this .1% in our DNA, in order to give you your results. When a testing company receives your sample, they compare your pattern of SNPs to different reference populations in their database. These reference populations contain SNPs known to exist more frequently in different modern populations in the world. Then the testing company will give you a percentage that represents how strongly your pattern of SNPs resembles that group. But this process has a bunch of important limitations and this is where things get complicated. Lots of markers are found in multiple populations around the world. First, even trying to classify humans into groups in the first place is tricky. Human genetic diversity isn’t organized neatly into groups like countries or continents. Take a look at the distribution of this SNP that affects how a person absorbs folic acid. It’s commonly found in Mexico, but also in Chile, or even China, just as often. So let's say that a particular marker is found in the South Asian population 30 percent of the time. There's still a possibility that when you inherited this marker you got it not from somebody who was South Asian, but from somebody who was in some completely different group that also happened to have that marker. Second, testing companies put together their reference populations based on academic research and other people that have taken genetic ancestry tests. And most testing companies aren’t clear about how many people are represented in their reference populations. So each company might have different reference databases, which helps explain why you might get different results from different companies. So what does this all mean for my results? This is a probability with a margin of error. So it's not that you overall are eighty-five percent West Asian, but that the particular spot that they happened to look at, eighty-five percent of those locations are associated with Western Asia in their reference population. So what about these other results? Am I really 2 percent African? You've got a lot of, you know, sort of small trace percentages here. Percentages that small are really not meaningful, again because that could be affected by having one person in the database. And if that one person gets reclassified later on because they get a larger sample, that percentage will disappear. Ultimately, DNA ancestry tests are really just giving us a probability, the testing company’s best guess. And that uncertainty isn’t made very clear in the results. Buried in my results I found this “confidence slider.” It turns out, my results were presented at about 50% confidence by default. When I increased it to 90%, my results got much more vague. All of a sudden I was "broadly" West Asian and a lot of my genetic markers were unassigned. So, DNA ancestry tests don’t actually tell us where our ancestors lived - they're really just giving us probabilities of where we’re likely to have relatives today. But so what if people misinterpret their results? Well that has consequences. They can make us believe that our ethnicities have these bright-line distinctions between them, like in a pie chart. When people are presented with test results and these percentage breakdowns and they are led to think that these tests can tell you your race or they can tell you who you are, that that leads to a way of thinking — makes us feel that there are very stark and clear biological differences between races. One study found that DNA ancestry tests reinvigorate age-old beliefs in essential racial differences, that our socially constructed racial categories like “white” or “black” are essentially different from each other. Some groups have even turned to genetic ancestry tests to try and prove their “racial purity." DNA ancestry tests can be useful. Search YouTube and you'll find hundreds of stories of people using them to find lost relatives and fill in their family histories. And, to people who don’t know a lot about their ancestry, the tests offer the best available estimate. But it’s important to remember that, despite their marketing, these tests are just a company’s best guess at matching your genetic markers to different parts of the world. What they’re not going to tell you is whether you should wear a kilt or not. DNA ancestry tests might not be as informative as you want them to be, but more and more people are still taking them. And this giant database of genetic information is becoming super valuable to an unexpected group: Law enforcement. We’ve teamed up with Verge Science, to look into how your privacy is at risk because of genetic ancestry tests, even if you’ve never taken one.
Info
Channel: Vox
Views: 1,488,168
Rating: 4.8041625 out of 5
Keywords: DNA, dna test, DNA sequencing, Vox.com, vox, explain, explainer, dna testing, ancestry.com, what are dna tests, dna tests explained, how do dna tests work, gene sequencing, genetic testing, ancestry, family tree, family history, 23 and me, are dna tests real, what do dna tests show
Id: IIWlatQt4KE
Channel Id: undefined
Length: 7min 24sec (444 seconds)
Published: Tue Apr 16 2019
Related Videos
Note
Please note that this website is currently a work in progress! Lots of interesting data and statistics to come.