Dad’s Ethnicity Estimates on AncestryDNA vs Family Tree DNA: Robledo, Nieto, Salas, Jimenez

I have blogged about the long-time brick walls I faced for almost 15 years with my dad’s family history, both on his father’s Robledo paternal line (hence the one-name study I just started last month), and on both of his mother’s lines. If you are a regular reader, you know that in May 2013 I had a major breakthrough on Dad’s maternal line, and have been on quite a roll since.

Venturing into DNA

Because of these brick walls, particularly his main Robledo paternal line, I had Dad take an AncestryDNA autosomal test (DNA from both sets of parents, all ancestors) back in January 2014, in hopes of identifying others researching his same lines. The results were processed in March 2014, and I have occasionally checked back to review new matches, but never really did anything with these results because I didn’t really know much about genetic genealogy. So at last month’s combo RootsTech and FGS conferences, I attended as many DNA sessions as possible, and felt equipped to now start doing something with Dad’s results. I say something, because I am still a total newbie at this.

The first thing I did — literally, the morning after I flew back home — was download Dad’s raw AncestryDNA data and transfer it to Family Tree DNA (FTDNA) to get a second analysis run against the data, and a second database of genetic matches. Unfortunately for me, FTDNA was in the midst of big time computer glitches, and it took them 3 weeks to process Dad’s results, then almost another 3 days for them to unlock those matches after I paid for that service. This process likely would have taken longer had I not repeatedly called FTDNA, retrying after many busy signals, and sitting on hold for long periods of time. All his FTDNA data finally became accessible to me this past week!

I will delve deeper into this journey, but the first step was a comparison of how each company analyzed Dad’s ethnic origins.

Ethnicity Comparison

What Dad Thought

Dad and I always knew about his Mexican origins; his paternal grandparents immigrated from there in the 1910s. He had always heard that his mother was born in Arizona, but we assumed Mexican origins for her as well. Consequently, we assumed Spanish ancestry for both of these lines. Don’t all Hispanic surnames trace back to Spain at some point? Maybe not, I don’t know. We are both highly confident that his ancestors did not all have pure Spanish blood. Like most Spanish New World families, those who might have had pure Spanish or even European blood had to have ultimately married into the indigenous populations, producing Mestizos.

The other thing Dad always heard is that the Robledo surname had some Sephardic Jewish blood, which seems possible due to the Sephardim presence on the Iberian Peninsula.

What DNA Tells Us

The analyses from AncestryDNA and Family Tree DNA are pretty similar, which is reassuring. They just label and break down ethnic origins a bit differently. Ancestry also provides estimates for much smaller sub-regions. Although from what I learned at RootsTech and FGS, estimates at these smaller regional levels are not considered very accurate.

Both identify European origins as Dad’s primary ethnicity. Ancestry breaks this  into two sub-regions: the largest being the Iberian Peninsula (Figure 1), but Italy/Greece (Figure 2) comes in close behind. Iberian Peninsula is no surprise. We expected this, based on Spanish assumptions. The high estimate for Italy/Greece came as a surprise though, since we have never heard of Dad having ancestors from this region. However the close proximity of these two regions as well as the Roman conquest of the Iberian Peninsula could account for this. FTDNA’s Southern European region includes both the Iberian Peninsula and Italy/Greece (Figure 5).

The Scandinavian part of those European origins is a big surprise. Although each company only estimates Dad’s DNA to have a possible trace amount: FTDNA at a 5% estimate (Figure 6) and Ancestry at less than 1% (Figure 4).

Not surprisingly, both companies rate Native American (indigenous peoples) ancestry as Dad’s second largest ethnic region of origin, in the New World (the Americas). What is surprising, is their discrepency in numbers. Ancestry estimates 33% (Figure 3) while FTDNA only estimates 21% (Figure 7). But this discrepancy might be due to what geographic areas each company counts as Native American. FTDNA (Figure 7) only classifies western Canada, the western U.S., Mexico, Central America, some of the Caribbean, and the small northwestern portion of South America as Native American. Ancestry classifies Native American as most of Canada and the U.S.; all of Mexico, Central America, and the Caribbean; and all but the southern tip of South America (Figure 3). FTDNA classifies less than half of New Mexico as Native American in origin (?!). Maybe I am interpreting these maps wrong? Perhaps our New Mexico ancestors just had more European ancestry than our California or Mexico ones.

The African, Asian, and Middle Eastern estimates get harder to interpret because the estimates are much smaller, and because each company classifies these differently. Ancestry puts North Africa as a sub-region under Africa, while FTDNA puts it as a sub-region under Middle Eastern. Yet Ancestry identifies the Middle East as a sub-region of West Africa.

What about Dad’s hunch that he has Sephardic Jewish ancestry? FTDNA either does not provide estimates at this level, or Dad didn’t even rate at less than 1% with them. Ancestry estimates that Dad has 3% European Jewish ethnicity in his DNA. No clue at this point if this is from his father’s side, or his mother’s side. No idea if this is Sephardic, Ashkenazi, or a mix of both since Ancestry does not make such a distinction. However, since his main regions of European ethnicity were historically home to Sephardim, I will assume Sephardi ancestry.

AncestryDNA

Following are the visual representations for how AncestryDNA interprets Dad’s ethnicity. See this blog post from Ancestry.com to learn how they estimate ethnicity.

Figure 1: Iberian Peninsula ethnicity estimate. Click image for a larger view.

Figure 2: Italy/Greece ethnicity estimate. Click image for larger view.

Figure 3: Native American ethnicity estimate. Click image for a larger view.

Figure 4: Ethnicity estimate overview. Click image for larger view.

Family Tree DNA

Following are visual representations of how FTDNA interprets Dad’s ethnicity.

Note that FTDNA shows you Family Finder Matches, genetic cousins in their database, who share your ethnic origins for your top three ethnic groups.These get displayed on the maps as Shared Origins (I have blocked out the names and faces of my matches, to respect their privacy). FTDNA members must opt-in to allow matches to see their ethnicity as Shared Origins. AncestryDNA does not provide this type of visual aid for genetic cousins and ethnic origins; you have to open up each individual cousin match to see their ethnicities.

Figure 5: FTDNA ethnic makeup overview. Click image for a larger view.

Figure 6: European ethnic makeup. Click image for a larger view.

Figure 7: New World ethnic makeup. Click image for a larger view.

Figure 8: East Asian ethnic makeup. Click image for a larger view.

Figure 9: Middle Eastern ethnic makeup. Click image for larger view.

Figure 10: West African ethnic makeup. Click image for a larger view.

What Next?

Now comes the hard part with the autosomal results, reviewing all the matches in both AncestryDNA and FTDNA to try to find common ancestor(s). I have already uploaded the raw autosomal data to GEDmatch, but need to learn how to actually use that tool.

I have also ordered a Y-DNA test from FTNA for Dad to further trace his patrilineal line. And although it won’t help me trace his maternal line, I have an mtDNA kit to test one of his paternal female first cousins, which will help me trace Dad’s paternal grandmother’s matrilineal line. I still need to identify someone who could do matrilineal testing on Dad’s mother.

And I need to start on Mom’s DNA testing, of course!