Deciphering DNA’s Astrology: Admixture Tools

If your kit isn’t already uploaded to Gedmatch, by all means add it to their free database.  Not only can you compare yourself to testers from all the Big Three DNA companies, but there are numerous tools to explore, including admixture tools.  Even if you are lucky enough to ‘know’ where your family is from, the first portion of results many testers look at is their ethnicity charts, and Gedmatch’s admixture calculators really let you dive in to your ancestry.

When using any of the Admixture, or Heritage/Ethnicity tools, it is vital to understand that they are only estimates of your ancestry.  It is also important to understand that certain components are expected across populations. For example, in Figure 1, we have a group of Irish testers.  Each kit has been run through Eurogenes K13, and the results are displayed in the table.  The circled column to the far right is the average result for a native Irish tester.  As you see, these Irish testers all display a small amount of Native American ancestry despite not having Native American ancestors.  The small percentages are merely an indication of deep ancestry, or DNA that was in the Native American population prior to traveling to the Americas. So, in my grandmother Gladys’s case, even though she has almost 2% Native American according to the calculator, it can be disregarded as not recent.

Figure 1

Another important factor to keep in mind is that different tools give different results, based on the populations that are sampled and the populations that the tools are intended for. For example, my grandmother Leona gets results that best fit what we know of her background with the MDLP tools, as she has North Eastern European ancestry. Eurogenes  is the go-to for my Western European and British Isles grandmother Gladys.  Were I predominately of African origin, I would use the Ethiohelix tool, and if I were Euro-African mixed I would gravitate towards the puntDNAL tool. Harrapaworld is for South Asians.

Currently, I have tested with both Ancestry and FTDNA, but my preference for admixture tools remains with Gedmatch because of the flexibility and additional ‘Oracle’ tools.  Depending on the option you choose, Oracles make a ‘best guess’ at the origin of your parents/grandparents. The closer the distance to the population, the more affinity, or more alike your DNA is.

My Ancestry results are slightly more exotic than FTDNA. My FTDNA results are limited to Western and Central Europe, Scandinavian, and Eastern European.  Given the differences in the estimates, it is easy to dismiss the entire matter as cocktail party conversation.

However, I have tested some of my extended family, and one of my son’s also has extended family beginning to test.  Since we have testers from multiple generations, we can wring a little more information out of the calculators, and have a better idea of what to ignore and what to note.  Plotting the results from various admixture tools we can look for patterns, consistency, and anomalies  across the generations and within known relationship groups.

Figure 2

In the above figure, you can see two separate family groups.  The larger box to the left is my extended family and the smaller box to the right, and above it includes my son Jeremy’s tested relations.

Gedmatch has added a new admixture tool to its roster, Gedrosia DNA. Currently, there is a post tracking results at Anthrogenica.

“This calculator is most accurate for individuals with predominantly S Asian or W Asian Ancestry. It is least accurate for individuals with predominantly African or Native American ancestry. Since I have not used African populations to source allele frequencies, Africans will appear predominantly SW Asian”

No one in my test group is predominantly S Asian or W Asian, but like any other Europeans we have deep W Asian ancestry as well as a mtDNA haplogroup born in Pakistan.  Equally tantalizingly, my grandmother Gladys consistently throws up about double the amount of South Asian as is normally found in an Irish person. I was curious to see what Gedrosia would make of our DNA as a group.

Figure 3 Gedrosia K11 Results

As you can see from Figure 2, Gladys is my paternal grandmother, Steve is my father, and Jeremy and Gavan (half brothers) are my sons.  Leona is my maternal grandmother, Jackie is her daughter and my full aunt.  Diane is my (untested) maternal grandfather’s daughter, and Jackie’s half sister.  Our results are falling into line with other Europeans reporting their findings at Anthrogenica, with a few exceptions:

• My son Gavan returned an out-of -left-field 1.9% Indo-Chinese result.  His Irish born father is currently untested
• My Irish grandmother and half-Irish father have much less SW Asian than the rest of the group.
• My grandmother Leona has more East Asian than the rest of the group. The East Asian population sample are the Ulchi, an indigenous group from the far east of Russia. Leona had known Eastern European ancestry.

Gedrosia K11 This calculator’s 11 components peak as follows:

1- WHG (W European Hunter Gatherer) – Loushbour & NE Europeans
2- S Indian – Various S Indian tribal populations, such as Hakkipikki and Nihali
3- Gedrosian – The Baloch, Brahui, and Makrani of Pakistan
4- SW_Asian – Saudis, Yemenis, and Bedouin
5- Siberian – Nganasans
6- EEF ( Early European Farmers) – LBK, Sardinians, and Stuttgart
7- E Asian – Ulchis
8- Caucasus – Georgian, Abkhasians, Adygei, and Balkar
9- Kalash – Kalash of Pakistan
10- Indo-Chinese – Kusunda peoples
11- SE Asian – Ami & Dai.

The second tested family group includes myself and my son Jeremy, as well as Jeremy’s dad’s half brother Anthony/AJ and AJ’s mother Nancy.  Most of the figures in this group tell us what we already have guessed:  our results are standard for Europeans.

Since Jeremy’s dad James is deceased, as is James and Anthony’s dad Jim, we are unable to test them. We haven’t yet gathered enough DNA to recreate their genomes via Gedmatch’s Lazarus tool. Still, there are a couple of oddities to take note of:

• I have no Indo-Chinese and Nancy has very little, yet both AJ and Jeremy have small amounts which they would have derived paternally
• AJ has nearly double the amount of Nancy’s SW Asian, which he would have inherited paternally

What can be done with these results to further your genealogy?  Very little, really, as they are more a matter of interest for the curious.  Admixture calculators are being continuously created and will continue to be improved as population samples are added and calculators are refined.

Meanwhile, as long as you remember a few key points enjoy experimenting with the calculators and see if they match what you know about your family:

• Results are estimates and vary
• Use the right tool for your background
• Look for patterns and consistency in results. Make a note of anomalies and note if they also appear in your the results of relatives