North African Ancestry in the (Spanish) Caribbean

Although our North African ancestry estimates (based on Ancestry DNA's results) are low (1% now, but 2% before the update), it was interesting to see that we have DNA matches with actual North Africans. In addition, we also matched three people of Jewish origin who share North African ancestry (but lacked Spain and Portugal). This does support the idea of inherited North African/Berber origin inherited through our Hispanic side (via the Spanish Caribbean). In our case, this could be via the Canary Islands and the Iberian Peninsula, although we have not yet documented any Canary Island ancestry which might explain our North African admixture. In fact, it could just as likely be a result of the early Iberian population in Puerto Rico including more people who harbored North African ancestry in the 1500s and 1600s. Or, perhaps, the presence of "esclavas blancas" and other captives of morisco and/or North African descent brought to the Spanish Caribbean in the 1500s may have introduced another avenue for North African ancestry. 

Looking at our specific matches with North Africans (and Sephardic Jews, or their descendants, who matched us based on Northern Africa in Ancestry DNA's Compare Regions feature) we suspect a closer tie to the Maghreb rather than Guanches of the Canary Islands. This, however, is just a supposition based on our slightly closer shared cM of DNA with people of Sephardic ancestry. However, for transparency's sake, there is a small chance three of the North Africans we matched with could be the result of shared West African ancestry. One of them, North Africa 2, actually received 4% North-Central Nigeria, 2% Senegal, and 1% Yorubaland in the latest update. Therefore, there is a tiny chance we matched him via his small West African ancestry. Likewise, North Africa 3 harbored even more West African ancestry (20% Senegal, 7% Mali, 3% Yorubaland). Again, North Africa 4 also had West African ancestry (Senegal, Mali, Benin/Togo, and Yorubaland) so there's a small chance we are a DNA match via West Africa. Likewise, our shared 11 cM with North Africa 6 could also be due to their trace amounts of Spain, Portugal and Sardinia, each at 1%, in their estimated results. Anyway, if Ancestry one day allows its users to see exactly where on the chromosome they align with their matches, I could confirm it is via shared Northern African DNA that I match these two. Nonetheless, the match of 8 cM with North Africa 1, whose results are 99% Northern Africa, are indicative of a North African ancestry. In fact, North Africa 1's surname indicates a possible Berber background. 

To make sense of the matches with Jews, who, according to Ancestry, only share Northern Africa with me as a region, complicates things. With one of them, who we are calling Jewish 2, he is actually only 7% Sephardic Jewish. He is overwhelmingly of Askenazi Jewish origin (more than 80%), yet we share 9 cM and the only common region on Ancestry is Northern Africa. The other case, Jewish 1, is a slightly closer match (10 cM). Like Jewish 2, he has no Spain or Portugal in his ethnicity estimates, but a 1% Northern Africa. However, he is 39% Sephardic Jewish (plus 26% Ashkenazi). Jewish 3 harbored even more Northern Africa ancestry according to her estimates, but we only share 9 cM. Jewish 3, however, is 43% Sephardic Jewish. We find it interesting that our closer North African matches are often with Jews, perhaps hinting at shared mutual ancestry that goes back in time to medieval Iberia and North Africa. In fact, Jewish 4, who appears to hail from the same family as Jewish 1, was another distant DNA relative. Indeed, through the paper trail, we suspect some distant ancestors were Jews who converted to Catholicism. In fact, we have 3 additional distant matches with Jews on Ancestry who appear to be of (at least partial) Sephardic ancestry, but not always showing Northern Africa and/or the Iberian peninsula regions in their admixture results. 

Of course, to actually confirm this would require using chromosome painter and a more extensive search for potentially bigger matches (8-10 cM is still small and there's a chance for false positives). It would also be worthwhile to compare results on Gedmatch to ancient DNA samples from North Africa or with Guanches from the Canary Islands. Alas, the closest we have come to that is the sub-par Hunter-Gatherer estimate on Illustrative DNA, which assigned us 1.6% North Africa. These are the next steps, particularly for making sense of how North African ancestry is part of the story of not only Spain, but Latin America.