”DNA Doesn’t Lie”: How To Accurately Interpret DNA Test Results
When you get your DNA test results, there's a good chance your DNA matches aren't as straight-forward as you had hoped, and there's a lot of room for misinterpretation. Here's some helpful things to consider as you begin your journey into the world of genetic genealogy!
If you’ve gotten involved in genetic genealogy at all in the last few years, you have probably heard the phrase “DNA doesn’t lie.” This is meant to imply that, while records left behind by our ancestors can sometimes be incorrect, falsified, or not necessarily representative of biological relationships a person cannot change their DNA, and hence cannot leave behind an incorrect or falsified genetic record. While this is true, that doesn’t mean DNA test results are always clear-cut and easy to interpret. If we aren’t careful in our analysis, we run the risk of misinterpreting the results and coming to the wrong conclusion. Let me illustrate this with a few examples.
Using DNA Test Results To Reveal Biological Parents
I had a client some time ago who I’ll call Jill. She was told growing up that her father did not believe she was his biological child. His reasoning for this belief was that her older siblings had all had relatively easy births, but Jill’s mother had a much harder labor with her. This bothered Jill, obviously, and she eventually decided to answer the question through DNA testing. At the time she performed the test, her parents had both passed, but she did have a paternal aunt still living who agreed to test as well.
When the results came in, the testing company predicted Jill’s relationship to her aunt was that of a “half-sister.” When she saw this, Jill was understandably upset, and began questioning whether her father was right after all. However, Jill was unaware of genetically equivalent relationships, where different genealogical relationships can share equivalent amounts of DNA. It’s important to remember that, no genealogical relationships can be definitively identified using only the amount of shared DNA. All other relationships require additional information. Even parent-child relationships need to be considered within the context of which tester is older and whether the parent had an identical twin or not.
For other relationships, there are often several possibilities for a given amount of shared DNA. For example, a full aunt, a half-sister, a grandparent, and a grandchild all share an average of 25% DNA. Fortunately for Jill, she had additional relatives on both her maternal and paternal sides who had tested. Analysis of the amounts of DNA she shared with those relatives indicated her paternal aunt really was her aunt, and that the DNA testing company had reported the wrong relationship from a list of possible relationships. She was finally able to assert that her father really was her biological father all along.
When Shared DNA Amounts Don't Match
In another case, a client who I’ll call Leslie came to us for help in identifying her biological father. We began sorting her matches into maternal and paternal matches (which was relatively simple, as she had a number of close maternal relatives who had already tested), and then sorted her paternal matches into related groups. Once we had put her paternal matches into family groups, we tried figuring out where Leslie might fit in. However, try as we might, nothing really worked – either she shared too much DNA for this relationship, or too little DNA for that relationship. It was obvious something was missing that could explain what we were seeing, so we began exploring the family trees of these matches beyond their most recent common ancestors.
We eventually found the family group with her closest paternal matches had one branch of descendants that married another branch of descendants. This marriage led to the descendants of that marriage sharing an inflated amount of DNA compared to those who only had one line of descent in this family group. Once this double relationship was accounted for, everything started to make sense, and Leslie’s biological father was soon identified. But it wasn’t until that double relationship to one branch of descendants was found that her paternal line could be clearly identified.
If you find yourself in a similar situation, where the amounts of shared DNA between testers just aren’t making sense, you can try asking yourself these questions to see if you might be misreading or misinterpreting the results:
- What are the genetically equivalent relationships for this amount of shared DNA? The smaller the amount of shared DNA is, the larger the number of potential relationships is. Blaine Bettinger’s Shared cM Project and a corresponding analysis tool at DNA Painter is an invaluable tool in identifying these relationships.
- Does the amount of shared DNA make sense for the expected relationship? In addition to there being multiple possible relationships for most amounts of shared DNA, all relationships have an average amount of shared DNA, and a range of the amounts of DNA individuals in that relationship have reported to share. In Jill’s case, her aunt shared about 25% of her DNA, but a full aunt could share as low as 16% or as high as 30%.
- Do you or your matches have genealogies with any pedigree collapse or endogamy? Both of these can cause matches to share more DNA with each other than you might otherwise, which can make identifying your genetic relationship to some of your genetic matches more difficult.
- Are there other individuals on the line of interest with known relationships who have tested? If the tester’s relationship to one individual is difficult to determine, having other known relatives on that line to compare with can help identify where the difficulty may be coming from. For Jill, having multiple relatives on her father’s side in the testing database helped us confirm her aunt really was her aunt; if those other relatives had not already tested, that would have been our first recommendation.
While DNA may not lie per se, it will not always give a straightforward, clear-cut answer to your research question. Being aware of some of the challenges in interpreting amounts of shared DNA can help you correctly interpret the amounts of DNA you share with your matches, and clarify what at first glance might seem to be unusual or unexpected results.
Everyone seems to be running out of time, for work, for family events, for everything in general. As with any new research venture, there is a steep learning curve in learning how to work with and understand your DNA test results. While you can (and should!) do all you can to learn about them yourself, you may also benefit from getting help from those who can work with your results now. A professional can help you focus on what you want to learn first, show you what your test results say about it, and what you can do to further solve your research problem.
DNA test results are like an onion; the more layers you peel back, the more you find underneath. DNA test results present you so much information—ethnicity estimates, genetic relatives, shared match lists, SNPs, STRs, triangulated segments, and more. Yet for all this data, there is little instruction to help you make sense of everything you’re seeing. A professional genetic genealogist can sort through the data to find the information that matters to you, and help you interpret it correctly.
Over the last decade, DNA testing has proven to be a powerful tool in genealogy research. One of the strategies repeated most often is the recommendation to ”cast a wide net”, or test at all the 
height=”1100″ />
I recently had the opportunity to read The DNA Guide for Adoptees, by Brianne Kirkpatrick and Shannon Combs-Bennett. Shannon is a 
