Amber's Genealogy and Family History Blog: My AncestryDNA Journey Part V: AncestryDNA Update

On the evening of October 17 I went to Ancestry's website and saw that they had finally released their updated version of AncestryDNA. These are the genetic ethnicity estimate results I received:

As you can see, they are quite different from my previous results and much closer to what I would have expected. I was very excited to see Europe West appear in my updated DNA results, as it was entirely absent in my previous results, despite the fact that my maternal grandfather's lineage was largely Dutch and German. Seeing this I wondered, why did it show up this time when it didn't before? Several other questions came to mind as well, such as why do I still see percentages that don't necessarily make sense (e.g. the 7% Irish result)? What is the meaning of the range given to the ethnicity estimates and where did it come from? To answer these and other related questions, we have to take a closer look at the updated AncestryDNA.
First, let's examine the ethnicity estimates and their associated ranges. To estimate your genetic ethnicity, Ancestry compares your DNA to the DNA of a "reference panel". The reference panel consists of 3,000 DNA samples from people from 26 distinct global regions who not only are native to their particular regions, but also whose ancestors for several generations were native to that region. Ancestry is operating under the assumption that if a person's parents, grandparents, etc. lived in a particular region, then it is probable that their more distant ancestors also inhabited this region.
The way Ancestry calculates your genetic ethnicity in this updated version is they run 40 analyses on your DNA using randomly selected portions of your DNA. They compare your DNA to their reference panel, looking for genetic markers specific to a given region. Since different segments of your DNA may contain varying percentages of traits specific to a particular ethnicity, Ancestry's results may vary from one test to the next. So my DNA results from the 40 trials may have yielded results like this (note that this chart I just made up for illustrative purposes):

Ancestry then averages the results of the 40 trials, which is where the percentages come from. So, out of their 40 tests on my DNA, the average amount of Europe West genetic traits in my DNA was 34%. They create the range by figuring out upper and lower bounds wherein most of the results from the test fall, although I'm not sure of what computations they perform to determine the range. My range was 12%-56%, meaning that most of my results fell between these two percentages. On my fictionalized chart, these results could be represented in the following manner, with the black line giving the average result and the blue shaded region representing the range:

You could still have some tests with results that were a bit lower or higher than what the range covers, as is the case with my made up chart, but those results would be in the minority.
That's a basic overview of where the ethnicity estimate and its associated range comes from. Now let's take a look at some of the other features of AncestryDNA 2.0. Below the map which you see in the first image I posted, Ancestry provides a comparison of how my results compare to a person native to Europe West along with some examples of results for people native to Europe West, as shown in the picture below:

The first thing to notice about this is the size of the reference collection, that is the number of people Ancestry used to come up with their estimate. In this case, it's 416 people, which is not all that large, and in many regions Ancestry used an even smaller number of people to generate their results. Using a small number of people to determine the unique traits of a population can leave some room for error.
Another thing to note is that even people native to Europe West on average have more genetic markers from other regions (52%) than they do genetic markers specific to Europe West (48%). From the examples they provide of results for people in their reference collection, we can see that in some cases, people native to the region didn't have any of the Europe West Genetic markers or very few. This highlights where some of the difficulty with determining genetic ethnicity stems from. Although there are a few regions in the world that are fairly insular, most populations have intermingled over time, which can make it difficult to determine from people living today which traits are specific to a given region.
You may be wondering what other ethnicities are prevalent in the populations of a given region. The chart below the Genetic Diversity chart provides some insight:

If there are some other ethnicities often found in a group of people, then it seems plausible that traits could be misidentified or something like that, which could explain some of the unusual results. This just goes to show that genetic ethnicity is an imperfect science. Fortunately it is one that is constantly evolving as new techniques are developed and I am all in all pleased with Ancestry's update.

Photo of the Day

Lucy Clifton Biggs with brother Walter Joseph Biggs Jr.

Lucy Clifton Biggs

Lucy Clifton Biggs (on right) with cousin Kate Southall (see the Photo of the Day from this post for more on Kate Southall)

Lucy Clifton Biggs

Lucy Clifton Biggs was born February 9, 1880 in Norfolk, Virginia to Annie (Southall) and Walter Biggs (profiled in the Photo of the Day segment of this post). By the time Lucy was two, her family had relocated to a farm near the town of Big Spring (today known as Elliston) in Montgomery County, Virginia on land purchased by her paternal grandfather Kader Biggs. According to the 1940 census, Lucy attended four years of college. She married, probably between 1910 and 1920, Lewis W. Langhorne. However, by 1923 she had been widowed. She lived out the remainder of her years in Salem, Virginia. She died in Salem on September 14, 1967. She was buried in Sherwood Memorial Park in Salem, Virginia.

Walter Joseph Biggs was born on June 4, 1886 near Big Spring, Virginia. He attended the New York School of Art (according to the 1940 census, he spent one year there) and became a renowned illustrator, being inducted into the Illustrator's Hall of Fame (here's a link to the induction speech and here's a link to some of his illustrations. And another link, with some biography. Also, if you do a Google image search you will find all sorts of paintings of his). He married probably between 1920 and 1925 Mildred Armstrong, whom he divorced sometime between 1930 and 1940. Although he spent much of his adult life living in New York, he died in Roanoke, Virginia on February 11, 1968. He was also buried in Sherwood Memorial Park.

Sources for Lucy Clifton Biggs:

"Descendants of the Reverend Daniel Southall of Eastern North Carolina" by Seth Warner, The Genealogist Vol. 22, No. 2