Sunday, 18 March 2018 and the EDGC "Ancestry and Genealogy" App: Another third-party analysis tool

In an earlier blog post I wrote about the third-party DNA analysis tools, Gencove and GenePlaza. In this post, I will review my experiences with a new third-party tool: The EDGC "Ancestry and Genealogy" app, available at the website resembles GenePlaza in that you upload your autosomal raw data and then shop around for apps to analyze it. One of the available apps is "Ancestry and Genealogy", developed by the Eone-Diagnomics Genome Center (EDGC). The app costs $9.99, and gives you a print-out report in PDF format containing various types of admixture analyses.

Here is my admixture breakdown according to EDGC, based on my FTDNA raw data:

These results are roughly consistent with my testing company results: Mainly European with some small contributions from other continents. The intra-European percentages, however, are way off the mark; they seem random, and I don't think they should be given much weight. The non-European percentages, however, closely resemble the results I get from the GEDmatch calculators.

I should add that in other people's reports, I've seen percentages as low as 0.02 %, and it seems likely that the cutoff point is 0.01 %. This is extremely fine-grained and makes me eager to know the margin of error. However, this information is not provided.

The EDGC report includes a chromosome painting showing contributions from both parents. To my knowledge, the only other place you can get something like this is 23andMe. (GEDmatch also provides chromosome paintings, but with the chromosome pairs blended and not easily distinguishable).

My EDGC chromosome painting looks like this:

I like this. It looks quite mixed and colourful!

Not surprisingly, East Asian is my most significant non-European category (at 2.74 %). That I have some kind of East or Central Asian ancestry seems to be an established fact by now, since East or Central Asian admixture is picked up by every extant vendor and third-party tool I have used, with the exception of MyHeritage. Most of my East Asian segments are read as Scandinavian at 23andMe, so I wonder if these segments might represent ancient East Asian admixture that has become "embedded" in the Norwegian population as a whole. It might perhaps be a signal of Siberian admixture through Sámi ancestors.

South Asia is also a rather significant contributor to my genome ("Asia" in the percentage chart, at 1.38 %). However, due to the Indo-European migrations that tie Europeans and South Asians together genetically as well as linguistically, this South Asian admixture might in fact be a misreading for European (more specifically, Indo-European ancestry from the steppes in modern-day Russia). If any of the alleged South Asian DNA is real, it is probably connected with my Romani ancestry or my Danish colonialist ancestor(s?) in Southern India.

I don't really know what to say about the "Admixed American" category. The fact that it says "admixed" is enough for me to take it with a huge grain of salt, and most of my blue segments are probably European. If this category is supposed to be read as Admixed Native American (I get a total of 2.2 % Native American in the percentage chart), and some of it is in fact real Native American DNA, it is difficult to tell if it is real (recent) Native American DNA or if it is Siberian/Central Asian DNA that has been shared between Native Americans and Europeans since ancient times. Since I do get a very small amount of Native American at 23andMe, it is likely that some of my "Admixed American" reflects real (recent) NA ancestry. My Native American ancestors were almost certainly Taínos from the island of Hispaniola.

Next, the African. Although it is a small amount (0.51 %), it seems to be concentrated in a few significant segments (notably the one on chromosome 11, and the slightly smaller ones on 2, 3 and 7), which is one indicator of small amounts of admixture being real and not noise. The tiny African segment in the middle of chromosome 6 seems to be the same as the tiny East African segment reported by 23andMe. The remaining African segments are not reported by 23andMe, which suggests that they are mainly statistical noise. However, the fact that I have known (or, should I say, very strongly suspected) African ancestry opens up the possibility that at least some of those segments might be real. I do wonder about that long segment on chromosome 11!

Interestingly, the EDGC report's Single Population Sharing section indicates that I have a genetic similarity (albeit at the lowest detectable level, <1 %) to populations in Western, West-Central and Southern Africa. This is consistent with ancestry from Africans who were transported to the Americas as slaves. And yes, the Single Population Sharing analysis does have a 0 % category; I have seen at least one report with fewer shaded countries than mine has.

In sum, I would say that the EDGC app is interesting on a continental level. It has raised my curiosity as to whether my 23andMe results might be missing some African segments. On an intra-continental level (in this case, within Europe), I'd say this analysis is rather useless, at least for me. It might be more informative for people with majority ancestry from more genetically diverse continents such as Africa.

My favourite feature is the chromosome painting.

No comments:

Post a Comment