Here are some reasons why your estimated percentage for a region based on the AncestryDNA reference panel might be higher or lower than you thought likely:
It could be due to the genetic influence of neighboring regions.
Over the last thousand years, there have been times when some groups of people were isolated from neighboring populations. Isolation gives populations a chance to develop a unique genetic signature.
When individuals from two or more previously separated populations begin intermarrying, the previously distinct populations become more difficult to distinguish. This combination of multiple genetic lineages is called admixture. Regions that border each other are often admixed — sometimes to a great degree.
For example: we find that most of the people on our reference panel in Spain have about 13% of their DNA that matches to the Europe South region. This could be due to relatively few cultural and trade barriers between the two regions, as well as conflicts between the regions over the last thousand years.
The chart below shows the average admixture between European countries (see this chart for all 26 regions based on the AncestryDNA reference panel).
It could be that the estimate is on the edges of our predicted range
For each region based on our reference panel, we show an ethnicity estimate as a percentage, along with a range and a confidence level. The range is shown by a dotted black line:
The percentage in this example shows that this person most likely has 55% of a given region, but they could just as easily have a bit more or a bit less. In this example the range is between 40% and 70%.
When we calculate your estimate for each range, we run samples of your DNA through our comparison engine 40 times and use that data to create a probable range. So, in this example, the actual ethnicity from this region could range from 40% to 70%. (Learn more about how we calculate the range).
It could be due to the random nature of genetic inheritance
DNA is inherited in a random way. For example, if you have siblings, you got different pieces of DNA from your parents than they did. So you might have a different ethnicity estimate than your siblings, or anybody else in your family, just because of the random way that pieces of DNA are inherited.
Possible ethnicity estimates for other siblings:
Here's another example. If you have a great-great-grandparent with Native American ancestry, you might expect to have 1/16th (about 6%) Native American ancestry. However, the pieces of DNA that you inherited from this great-great-grandparent are random. When the DNA was passed from your great-great-grandparent, to your great-grandparent, to your grandparent, to your parent, and then to you, some pieces of DNA from this great-great-grandparent were not passed on. Since you might not have much DNA from that great-great-grandparent, you might not show up as having any Native American genetic ethnicity. This doesn’t prove that you don’t have any Native American ancestors or heritage. It just means that you didn’t inherit the genetic markers AncestryDNA recognizes as indicating Native American ethnicity.
So, keep in mind the random nature of DNA inheritance, especially if you don't have exactly the genetic ethnicities that you expected.
It could be because ethnicity estimation is still an open problem
Another thing to keep in mind is that ethnicity estimation is a problem that scientists around the world are actively working on. At AncestryDNA, we are on the cutting edge of this science, and we do our best to give you the best possible estimate. However, the fact is that ethnicity estimation is hard.
There are a number of reasons why ethnicity estimation is difficult. If populations remained isolated all the time and never intermixed, it would be easy to find out what ONE ethnicity each person belongs to. History shows that this is simply not the case. After centuries of migration, military invasions, intermarriage, and such, most people today are a mix of different ethnicities.
You can be confident that we are giving you the most accurate estimate possible given our current dataset and knowledge of human genetics. However, you can also be sure that we are working hard to continue to improve the resolution and accuracy of your ethnicity estimates.
Still curious to understand more? Cool--we're glad you're as interested in genetics as we are. Check out our white paper on ethnicity prediction.