Latin American genetic structure (Native, Euro, Converso & SSA clusters)

[Tomenable]

Fig. 1. shows proportions of Euro and Native admixtures in several countries:

Study: https://www.biorxiv.org/content/early/2018/01/23/252155

PDF: https://www.biorxiv.org/content/earl...52155.full.pdf

Supplements: https://www.biorxiv.org/content/bior...1/252155-1.pdf

Historical records and genetic analyses indicate that Latin Americans trace their ancestry mainly to the admixture of Native Americans, Europeans and Sub-Saharan Africans. Using novel haplotype-based methods here we infer the sub-populations involved in admixture for over 6,500 Latin Americans and evaluate the impact of sub-continental ancestry on the physical appearance of these individuals. We find that pre-Columbian Native genetic structure is mirrored in Latin Americans and that sources of non-Native ancestry, and admixture timings, match documented migratory flows. We also detect South/East Mediterranean ancestry across Latin America, probably stemming from the clandestine colonial migration of Christian converts of non-European origin (Conversos). Furthermore, we find that Central Andean ancestry impacts on variation of facial features in Latin Americans, particularly nose morphology, possibly relating to environmental adaptation during the evolution of Native Americans.

Allele-based analyses have previously documented that broad patterns of Native American population structure are detectable in admixed Latin Americans6,7. SOURCEFIND analysis extends these results by enabling the inference of 25 Native American ancestry components across Latin America, resulting in a high-resolution picture of Native variation in the region (Figures 1B and 2A) and emphasizing the “genetic continuity” of pre-Columbian and admixed populations across the Americas. In addition, SOURCEFIND distinguishes between closely-related ancestry components from the Iberian Peninsula, as well as from the East and South Mediterranean (including individuals self-identified as Sephardic; i.e. Iberian Jews). The distribution of European ancestry in the CANDELA sample shows a sharp differentiation between Brazil and the Spanish American countries (Fig. 1C). In Brazil the predominant European sub-component matches mostly the Portugal/West-Spain reference group while in Mexico, Colombia, Peru and Chile mostly Central/South-Spanish ancestry is inferred (Figures 1C and 2B). This differentiation matches the colonial history, Portuguese migration having concentrated in Eastern South America while the Spanish settled mainly in Central America and Western South America1. The relatively small contribution inferred for the Basque and Catalan agrees with historical information documenting that Spanish migrants to America originated mainly in Southern and Central Spain8.

Also significant Italian and German admixtures in Southern Brazil were detected.

[Angela]

I debated whether to post this study. I decided not to because I'm really not ready for another confrontation, but whatever...

I disagree with Razib Khan's take on this with regard to the Conversos, and think he totally mis-characterizes a study done a few years ago actually investigating these claims of "converso" ancestry.

See:
https://www.gnxp.com/WordPress/2018/...medium=twitter

https://www.theatlantic.com/magazine...n-jews/378454/

I am not convinced that this additional "eastern Mediterranean" in New World Latinos is necessarily "Sephardic" in origin. There's nothing particularly distinctive about Sephardic genetics, in contrast to Ashkenazi genetics, so it's very difficult to pinpoint it. More "eastern Mediterranean", whatever they actually mean by that, didn't necessarily enter the Spanish New World genome through Jews. It sounds to me like typical sloppy Hellenthal thinking. As for Khan, when he was at FTDNA he made some snap judgments about that as well, and maybe he feels the need to defend it.

The reality is, however, that the majority of the first Spanish settlers in the New World came from Extremadura and surrounding areas. This couple of percent "extra" and more "eastern" ancestry could be the result of other migrations into Extremadura as well, and then onto the New World, not necessarily only by the Sephardics.

Not that I deny that some Sephardim might not have fled to the New World hoping to put distance between themselves and the Spanish Inquisition.

This leads me to the careful investigation of the claims of "Hispano" Americans in the southwest about so called "Sephardic" practices which can be found at my second link. If you read the article carefully, it's clear that some of these claims are nonsense. Many of the practices detailed by these supposed descendants of conversos are Ashkenazi in origin, and no Sephardim would have practiced them or passed on knowledge of them. Others were practiced by people who had converted to a Protestant sect which practiced certain "Jewish" customs. It's no coincidence that a lot of those claiming such ancestry have this kind of Protestantism in their history. In addition, it is absolutely true that a lot of Hispanos have attempted to distance themselves from Mexicans, on the grounds that they have no Amerindian ancestry. Jewish ancestry is acceptable, but not Amerindian ancestry. It reminds me of the groups like the Melungeons, who claimed for years that they had Jewish and Amerindian ancestry, anything rather than acknowledge the reality, which was that they were a mix of "Anglo" and "black".

That doesn't mean that some of the claims aren't valid, as I said. It just means it isn't as widespread or extensive as is claimed.

[Tomenable]

A few of my observations:

1) Brazilian European input is a mix of Portuguese, Italian and German. It is expected since 80% of the samples come from Rio Grande do Sul. I'm surprised by shortage of Polish ancestry - but most of Polish-Brazilians live in Santa Catarina and Parana, not in Rio Grande do Sul, where their samples come from. Another factor is that their European references do not include any Polish-like populations (they included only typical North-West Euros and some Finno-Ugric groups - Estonians, Finns and some Ugric group from Russia).

2) Colombian sample has a surprisingly low % of SSA. Likely because they sampled the Paisa region, which seems to be Harnizo, while most of Afro-Colombians live in the west, in the Caribbean region.

3) Their Mexican sample is 60% Amerindian, more than in most studies. I think the average for entire Mexico is ca. 50%. There is a cline of Native ancestry from the north (least) to the south (most).

4) I'm disappointed that there are no barplots or boxplots showing ranges of ancestries. Average/mean is one thing, but distribution is another thing. Some Latino populations have more variation, others have more homogeneous proportions of admixtures (e.g. Costa Ricans are all similar, Argentinians are diverse).

[Tomenable]

Their references for North-East Europe include only Non-Indo-Europeans: Estonians, Finns and some (probably Uralic-speaking?) group from Northern Russia, see Fig. 1.:



Why is Bulgaria counted as "Italian" reference?

[Angela]

Their references for North-East Europe include only Non-Indo-Europeans: Estonians, Finns and some (probably Uralic-speaking?) group from Northern Russia, see Fig. 1.:



Why is Bulgaria counted as "Italian" reference?

Bulgaria and Northern Italy don't plot far from each other, as Cavalli-Sforza knew thirty years ago, but you can tell them apart easily enough, so I don't think there's any excuse for that.

Also, a lot of Southern Italians went to Latin America too, so they're missing them by not having a sample from Southern Italy as reference.

If North Africa is included in their East/South Mediterranean, some of what they're picking up could more North African in the people who migrated to the New World, which would make sense if they came from Extremadura, Sevilla, etc.

[Sile]

Their references for North-East Europe include only Non-Indo-Europeans: Estonians, Finns and some (probably Uralic-speaking?) group from Northern Russia, see Fig. 1.:

Why is Bulgaria counted as "Italian" reference?

.
someone told me its due to the black sea peoples........north-italians and bulgarians have links with ostrogoths and gepids ..........but that would be after the collapse of the western Roman empire ....................I have never found a link, although Laz paper also links bulgarians with North_italians
.
not all goths are the same
.
The Ostrogoths were the eastern tribe of the Goths (a Germanic people) who rose in power in the area north of the Black Sea. The designation, Ostrogoth, taken to mean `Eastern Goth’, actually means `Goths glorified by the rising sun’ and gave birth to the term Visigoth (interpreted to mean `Western Goth’) by the Roman writer Cassiodorus who coined the term in the 6th century. Cassiodorus lived among the Ostrogoths and served their king Theodoric the Great. In an attempt to simplify the designation between the Germanic tribes which had moved toward the west, and those who remained in the east, Cassiodorus, deliberately or mistakenly, interpreted `Ostrogoth’ to mean `Eastern Goths’ and the others then became `Western Goths’ but the people themselves did not think of themselves along those lines. The Visigoths would, in time, accept and apply that term to themselves and the Ostrogoths had long known themselves by that name but neither tribe would have considered themselves `eastern’ or `western’ Goths.
The Goths first appear in history living in the area around the Black Sea. They made constant incursions against the provinces of Rome and proved a resilient and perpetual nuisance to the Empire until the invasion of the Huns in 375 CE. A large portion of the populace (according to some sources, 200,000) fled the area to seek the protection of the Roman Empire under the emperor Valens (these people became known as the Visigoths). The rest of the people remained, enduring the rule of the Huns, and retaining a certain degree of autonomy.
.
The earliest ostrogoths born in northitaly was Athal 230-295AD born in Verona , and the goths where still in northern Italy at 540AD when the Lombards came through

[Angela]

^^You just need to look at any PCA of modern Europe.

https://c1.staticflickr.com/4/3803/3...3e139a15_b.jpg



Also, when there are no North Italian samples, North Italians often get Bulgarians as their first match in Oracle type analyses.

[Hauteville]

Southern Brazilians as well as Uruguayans and Argentinians received lot of Italian immigration.

[Carlos]

The Spanish Jews could convert and stay in Spain, I do not think it was significant the march to America of these. 70% of the women who emigrated to the Caribbean were Andalusians. For this reason, the Andalusian way of speaking adapted to each country in the course of time is developed logically.

[Ygorcs]

Not many surprises coming from this study, if the supplements are a good indication of the main results. I think it's a pity that they relied so heavily on Southern "gaúcho"people for their analysis of Brazilian ancestry, since that is notoriously a very distinctive region of the country with its own history of settlement.

Nevertheless, I was surprised to see that Subsaharan admixture in some of the individuals from Rio Grande do Sul is a lot more than I'd expected, probably confirming that this state received more African influx than is often considered (another good indication of that is that, contrary to what most people think, the highest percentage of people who follow African-Brazilian religions is to be found in Rio Grande do Sul, not in the Northeast of Brazil or in Rio de Janeiro, where Subsaharan admixture is as a whole much higher).

As for their "Sephardic" findings, I also have my doubts that 100% of it is owed to Sephardic conversos, crypto-Jews and "cristãos novos" (cristianos nuevos) alike, but I'd say that their findings, coupled with the fact that they really established very specific regional admixtures, are very interesting, indeed, though not very surprising.

They point out that the "Sephardic-like" ancestry is significantly higher in Latin American than anywhere in Iberia among the many samples they analysed, so something must be found to explain that disparity between the core regions of European origin in Latin America and the actual result of that immigration in the New World. In one or another way those populations of Iberia that had more East Mediterranean ancestry contributed disproportionately to the gene pool of Latin Americas.

I'd say that some of that comes from people who had long become Christian or simply shifted back to Christianity in Central-Southern Spain and Southern Portugal as the political/cultural situation changed, but even generations later they could've probably been still less privileged by the new social structures than the northern people who resettled the southern areas of Iberia (the Reconquista wasn't this dramatic and unexpected event, it had more than 400 years to absorb people). I can definitely see that people with more North African or Middle Eastern ancestry could've been more easily lured into adventuring themselves in the New World.

But some of that Eastern Mediterranean admixture must've come with recent conversos or then with communities of people who had converted much earlier but were somehow more vulnerable and with less economic and military opportunities in Iberia due to the growing paranoia of "limpieza de sangre".

We have dozens and dozens of anecdotal records of conversos, new Christians and "moriscos" among the very earliest waves of immigration to South America (even the first fleet that "discovered" Brazil apparently had one Jew in its crew), including even Jews, those who hadn't even converted but were too useful to be discarded. There were also sizeable Jewish communities in the Northeast of Brazil, with even remnants of their synagogue in Recife before they were expelled due to their collusion with the Dutch against the Portuguese.

The sign of Sephardic ancestry that these scientists found is significant, but still very minor (in most cases less than 4%), so I don't think it's very unlikely to be true.

Remember that those "conversos" and "moriscos" would've been among the first to colonize the Americas, so they had much more time and were probably much more prone to polygamy (that was certainly the case among the first Portuguese who came to Brazil and were offered dozens of Amerindian wives in exchange for alliances, gifts and trade - the Native Brazilian tribes practiced this exogamy very intensely, by which they symbolically "adopted" the European husbands of their women, who thus could become "part of the tribe" - so-called "sistema de compadrio") than the later immigrants who arrived when the Americas were already well established colonies with stronger institutions. So I think it's definitely possible that even if only ~2% of the incomers were conversos, former Jews or former Muslims, their genetic impact would become even more relevant after several generations.

[Ygorcs]

The Spanish Jews could convert and stay in Spain, I do not think it was significant the march to America of these. 70% of the women who emigrated to the Caribbean were Andalusians. For this reason, the Andalusian way of speaking adapted to each country in the course of time is developed logically.

I really doubt the social and economic situation of former Jews and Muslims who stayed in France was very pleasant. I can definitely see why more of those who had known Muslims or Jewish past in their genealogy could have migrated more often to the New World with its attractive abundance of lands, possibilities and, most of all, lawlessness. Of course, we'd not be talking of a massive and concerted emigration wave, but a much higher tendency to be convinced to board on a ship because they'd have much less to lose in Iberia.

[Regio X]

Southern Brazilians as well as Uruguayans and Argentinians received lot of Italian immigration.

Other Brazilian states were also heavily settled by Italians. Sao Paulo is a good example. It received many Italians from almost all macro-areas of Italy, and they were easily integrated. One of the distinctive aspects of the (Northern) Italian colony in Rio Grande do Sul and Santa Catarina would be their relative isolation till few decades ago, which would help to explain why the language/culture was much better preserved in these areas. This would be true also for Germans, among others, with the difference that most of them really migrated to the Southern Brazil.

I took a quick look at this paper time ago. I'd have to see it closer, but if my memory serves me correctly, they tested people from the capitals of some states, and the capitals are not really representative of the upcountry in several cases. I could mention as example Porto Alegre itself (it received some migrants from the northern half of Rio Grande do Sul, of course, however, given its origin, I guess it represents better the southern half of the state, or more generally the area inhabited mainly by the so called "gauchos" - in the ethnic sense of the term, which is a general demonym as well), Florianopolis (originally Azorean, it would likely represent better the coast of Santa Catarina, while the upcountry in general is pretty different ethnically), Vitoria and others.

Edit: Sorry! I mixed up the study discussed here with the following:
Male Lineages in Brazil: Intercontinental Admixture and Stratification of the European Background
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4821637/

I haven't read the "Latin American genetic structure" yet; I don't know which pops they used.