PDA

View Full Version : Post-colonial pop. structure of N America. Finally a study I can literally relate to



MitchellSince1893
02-07-2017, 09:50 PM
Heber posted this in the new paper section. http://www.anthrogenica.com/showthread.php?709-New-DNA-Papers&p=212762&viewfull=1#post212762

I'm still digging through it, but as I'm 75% Colonial American mutt (mother is 100%), this is a quite fascinating genetic study of America in the last few hundred years.


Great paper available under Creative Commons licence.

Clustering of 770,000 genomes reveals post-colonial population structure of North America

Eunjung Han, Peter Carbonetto, Ross E. Curtis, Yong Wang, Julie M. Granka, Jake Byrnes, Keith Noto, Amir R. Kermany, Natalie M. Myres, Mathew J. Barber, Kristin A. Rand, Shiya Song, Theodore Roman, Erin Battat, Eyal Elyashiv, Harendra Guturu, Eurie L. Hong, Kenneth G. Chahine & Catherine A. Ball

http://www.nature.com/articles/ncomms14238

Abstract
Despite strides in characterizing human history from genetic polymorphism data, progress in identifying genetic signatures of recent demography has been limited. Here we identify very recent fine-scale population structure in North America from a network of over 500 million genetic (identity-by-descent, IBD) connections among 770,000 genotyped individuals of US origin. We detect densely connected clusters within the network and annotate these clusters using a database of over 20 million genealogical records. Recent population patterns captured by IBD clustering include immigrants such as Scandinavians and French Canadians; groups with continental admixture such as Puerto Ricans; settlers such as the Amish and Appalachians who experienced geographic or cultural isolation; and broad historical trends, including reduced north-south gene flow. Our results yield a detailed historical portrait of North America after European settlement and support substantial genetic heterogeneity in the United States beyond that uncovered by previous studies.

MitchellSince1893
02-07-2017, 10:32 PM
Excerpts from the supplementary info
Intact immigrant clusters
African Americans. Our interpretation of this cluster, identified in both the clustering of the subnetworks (second level of hierarchical clustering) and the spectral analysis, is strongly supported by the distribution of global admixture proportions (Supplementary Figs. 10, 11). The distribution of ancestral birth locations in this cluster closely traces the westward expansion of cotton cultivation and slavery, originating in the rich coastal plains of North and South Carolina, then progressing west until it reaches eastern Texas

European Jewish. One of the largest clusters initially identified in the IBD network (first level of hierarchical clustering) is the European Jewish cluster, which was also identified as a cluster in the spectral analysis. In the ancestral birth location maps for this cluster (Supplementary Figs. 19, 20), we do not find a large over-representation of ancestral birth locations at specific locations within the US. The strongest over-representation of Jewish birth locations is in or near New York and Chicago (OR > 2); these cities received large numbers of Jewish immigrants during the late 1800’s and early 1900’s.

Portuguese. In the second-level of the hierarchical clustering, we identify a cluster corresponding to the Portuguese (Table 1, Supplementary Data 2, Supplementary Fig. 19). The Portuguese began to immigrate to the US in large numbers in the late 19th century; immigrants were primarily men from the Azores and Madeira Islands, recruited to work on whaling ships 17. These men emigrated to the eastern US, establishing communities in various New England coastal cities (major regions include Providence, Bristol, Pawtucket in Rhode Island, and New Bedford, Taunton, Fall River in Southeastern Massachusetts); we do not find strong enrichment of ancestral birth locations of Portuguese cluster members in these regions, likely due to the influx of other groups to these regions. The Portuguese also immigrated to various cities in California, including the San Francisco-Oakland Bay Area, Santa Cruz, the Central Valley, and San Diego, and this is reflected in Fig. 3. In the mid-to-late 20th century, there was another documented surge of Portuguese immigration in America, mainly in the Northeast (New Jersey, New York, Connecticut, Rhode Island, Massachusetts).

Eastern Europeans. In the second level of the hierarchical clustering, we identify a cluster we call “Midwest immigrants” (Supplementary Data 2). Although we do not specifically discuss this cluster in the main text, we do discuss two of its stable subsets, Scandinavians and Finnish. This cluster is subdivided into additional large clusters in the third level of the hierarchical clustering (Supplementary Figs. 25, 26). Some of these clusters clearly relate to geographic structure and immigration patterns from Eastern Europe. For example, the ancestral birth location patterns of the "Eastern Europeans and Italians in Pennsylvania and Midwest" cluster (Supplementary Fig. 26) might correspond to the migration of nearly 24 million southern and eastern Europeans to the US between 1880 and 1920, before the restriction of immigration in 1924 18. Immigrants predominantly came from Hungary, Poland, Austria, Slovakia, Czech Republic and Italy between 1840 and 1870, with New York state and Pennsylvania the primary destinations for these immigrants.

Northern Europe (Finnish, Swedish, Danish, Norwegians and Scandinavians). Additional third level clustering of the "Midwest immigrants" cluster relates to regions in Northern Europe. A Finnish cluster was identified in the spectral analysis (Table 1, Supplementary Figs. 11, 19), and in the third level of hierarchical clustering (Supplementary Figs. 25, 26). Ancestral birth locations in the Finnish cluster and stable subset coincide with their historical record of migration to the US, and in particular to the Michigan Upper Peninsula 19. We also identify a stable subset in the Midwest immigrants cluster which we label as Scandinavians (Table 1, Fig. 3). Ancestral birth locations in this cluster closely correspond to the settlement pattern of Norwegians in rural Minnesota, North Dakota and Wisconsin, with large numbers moving to Minneapolis and Chicago

Irish. In the spectral analysis, we identify a cluster which likely corresponds to descendants of Irish immigrants. Six million Irish settled in the US in the 19th century, with immigration peaking in 1852 during the Irish famine 21. Irish migration was historically characterized by a highly localized pattern of chain migration, as migrants followed family members and neighbors to the same towns and cities in the U.S.21. In fact, almost half (46%) of Irish immigrants migrated to just 10 U.S. counties

Other European immigrant groups. In addition to the results already discussed, we identify substructure corresponding to other European immigrant groups when we subdivide the “Italians, Irish and Scottish” cluster (Supplementary Data 2, Fig. 26). For example, we identify a cluster whose ancestral birth locations are disproportionately concentrated in Scotland, Atlantic Canada and Ontario (Supplementary Fig. 26), corresponding to migration of large numbers of Scottish to Canada. Furthermore, we also find an Italian cluster (Supplementary Fig. 26). The ancestral birth locations for the Italian cluster are particularly concentrated in southern Italy, reflecting the predominant source of Italian immigration to the US.

Polynesians, East Asians and Hawaiians. In the first level of hierarchical clustering, we identify a cluster corresponding to Polynesians and East Asians (Supplementary Data 2) with only a small number of individuals, and thus we do not subdivided it further. The Hawaiian stable subset identified in this cluster is representative of the Polynesian population with some East Asian admixture. Reflecting this finding, Hawaiians have likely remained genetically isolated due to the large distance from the continental US, while Hawaii has a complicated history with recent, abrupt population changes and rapid growth

MitchellSince1893
02-07-2017, 10:39 PM
Supplementary Data on Hispanic Americans
Continental admixed groups
Mexico clusters. In the main text, we discuss the connection between the Northeast and West
Mexico clusters and Mexico-US migration patterns. We further note that areas that have not
traditionally seen a large influx of migrants to the US, such as southern Mexican states, are
poorly represented in the Mexican clusters (Supplementary Figs. 17, 18), reflecting our USbiased
sample.
New Mexicans. We identify a cluster corresponding to New Mexicans in both the clustering of
sub-networks as well as in the spectral analysis (Fig. 3, Table 1, Supplementary Data 2). This
cluster most likely represents descendants of the Nuevomexicanos, some of the earliest
European colonial settlers that migrated northward from New Spain along the El Camino Real
de Tierra Adentro trail25. Supporting this hypothesis, ancestral birth locations are
disproportionately found in parts of Mexico and New Mexico near this trail (Supplementary Fig.
20).
Puerto Ricans. In our discussion in the main text, although we do not single out Puerto Ricans
from other Caribbean Islands peoples, Puerto Ricans are by far the predominant Caribbean
group in our sample. Puerto Ricans typically have a mixture of European and African ancestry,
with smaller amounts of Native American admixture26, and this is reflected in our data
(Supplementary Figs. 9–11, Supplementary Data 2). In the second level of the hierarchical
clustering, we identify fine-scale structure on the island of Puerto Rico in the 3 largest clusters of
the Caribbean sub-network (Supplementary Data 2). These 3 clusters are clearly correlated with
geography of the island of Puerto Rico, as they roughly subdivide the island into 3 regions—
northwest, southwest and east (Supplementary Fig. 21). These 3 Puerto Rican clusters show
small differences in Native American, West African and European admixture proportions
(Supplementary Fig. 11) that only partially reproduce the findings of genetic variation across the
island of Puerto Rico26, perhaps due to differences in the composition of our database and their
sample. Interestingly, the Puerto Rican cluster has an enrichment of ancestral birth locations on
the island of Hawaii. This likely reflects the arrival of sugar cane plantation laborers to Hawaii in
the early 1900’s; Puerto Ricans have been documented as a separate ethnic group in Hawaii as
early as 1910, and they have constituted over 2% of the population up until 1950

Saetro
02-07-2017, 11:29 PM
German input was massive, but not represented. Probably needs subdividing.
Difficult if submitters still finding their way on this.
Some of my German matches just quote "Germany" or "Prussia" (nearly as broad) rather than province, for example Posen or Schlesien.
However, these latter were German/Slav mixed areas. Some who emigrated to Australia appear to have been mixed already and mixed further on arrival.

Cornwall is quite distinctive, genetically, according to POBI, and tended to have mining immigrants to USA who went to specific areas.
But submitters often just put "England" as origin, so maybe the further detail is not reliably retrievable.
Pity.

MitchellSince1893
02-07-2017, 11:31 PM
In case you were wondering what exactly "CEU" was and where it came from ...you know the Utah population sometimes seen in gedmatch admixture tools such MDLP (CEU), Dodecade (CEU) HarrapaWorld (utahn-white), puntDNAL (utahn-white) .

CEU=Utah Americans of Western and Northern European descent

13850

I sometimes get close admixture distances to CEU and Danish. Now I know why. CEU has a strong Danish affinity.

MitchellSince1893
02-09-2017, 04:49 AM
This is one of the many maps in the supplemental data.
13896

What it shows is that the ancestors of people from Coastal North Carolina were, for the most part, living in North Carolina 7-9 generations back.

This is exactly the same pattern I've seen in my own colonial ancestry.

Of my 128 5th grandparents (7 generations back), 96 of them (75%) were living in Colonial America, and all the ones I know about lived in either North Carolina (NC) or South Carolina (SC).

I've been able to find the birth and death locations for 68 of these 96 5th great grandparents.

These 68 5th great grandparents were born between 1732 and 1802 (Median year of birth is 1776).
42 were born in NC. Of these, 41 died in Eastern North or South Carolina and 1 died in Mississippi.
16 were born in SC and all 16 died in either Eastern South or North Carolina
4 were born in Virginia and 2 each died in NC and SC
2 were born in Pennsylvania, both died in NC
3 were born in Scotland, and all 3 died in NC
1 was born in Northern Ireland and died in NC.

So 58 out of 68 of my 5th great grandparents were born in eastern NC or SC, and 67 out of 68 died in either NC or SC, and one moved to Mississippi and died.

razyn
02-09-2017, 02:10 PM
I can relate, I'm also from the SE United States and have essentially all "colonial" ancestry. That sort of coincidentally aligns with the fact that autosomal DNA percentages fade with time. So, the modifier "post-colonial" in this article's title is a function of their DNA data source -- living people -- more than the paper trail data that accompany it. It doesn't really have anything to do with Colonial Dames, DAR, FFV, Mayflower Descendants, and similar genealogically based interests. Really, the chip tests are better at sorting the percentages in the Ellis Island alumni association -- which seems (anecdotally) to be well over half the US population, but it skews away from the southeast. The rest of us are too thoroughly blended, after ten or more generations in North America, for our short IBD segments to show much of a pattern of European source localities. In the Upland South region that includes German, Swedish, French, and other colonizing European populations that have long since blended with those from what the Romans called Brittania. In most cases they didn't fill out a lot of paperwork on arrival to these shores; so questions about which British shire, which side of the river Wye, which Irish county, French province, German state and so on are largely unanswered in the official records, this side of the pond. Family tradition is sometimes right, and official records are sometimes wrong, but in both cases it's a mess.

Calas
02-19-2017, 11:53 AM
What it shows is that the ancestors of people from Coastal North Carolina were, for the most part, living in North Carolina 7-9 generations back.

And interestingly enough just south of you, 700-1,100 Acadians arrived in South Carolina during the mid-1700s. I wonder how many people know that little bit of fact? I mean almost everyone thinks Acadians > Cajuns > Louisiana. Sorry to say that it wasn't that simple. They were settled in a number of eastern states.

And it is interesting, as Saetro said, that the German factor seems to have become rather downplayed. Wonder if that's due to intermarriage. Lack of actual knowledge such as Americans with a misconception of their ancestry claiming X on the sampling profile when they're really more Y.

AnnieD
02-20-2017, 08:46 PM
And interestingly enough just south of you, 700-1,100 Acadians arrived in South Carolina during the mid-1700s. I wonder how many people know that little bit of fact? I mean almost everyone thinks Acadians > Cajuns > Louisiana. Sorry to say that it wasn't that simple. They were settled in a number of eastern states.

And it is interesting, as Saetro said, that the German factor seems to have become rather downplayed. Wonder if that's due to intermarriage. Lack of actual knowledge such as Americans with a misconception of their ancestry claiming X on the sampling profile when they're really more Y.

I had no idea of the Acadian influence in South Carolina where I still have kin! B)

Growing up in 2 Colonial southern states, USA, only 1 of my family lines was deeply into genealogy and had a well-documented tree back to Scotland in 1100s (not sure how accurate but they certainly tried :P). However, even these kin did not routinely talk in terms of we English, Scottish or "Celtic" nor participate in ethnic associations such as Highland games or Ceilidh dances. When I moved to a major metropolitan area after college, I then discovered a revival of European ethnic associations such as Scottish Country dance, Scottish Highlanger games, Irish, Sons of Norway / Daughters of Norway Viking Festivals, Sons of Italy, etc.

In my family tree research thus far, I have uncovered one anglicized surname which appears to be more of a humorous failure at English spelling rather than deliberate attempt to look more Anglo. ;) However, there were certainly many instances of surnames changes to avoid discrimination or to assimilate more quickly.

AnnieD
02-20-2017, 09:02 PM
A partial Colonial (mostly British?) Mutt here, too! ;) I see in the Affiliations section at the end of the article two references to AncestryDNA. I am encouraged that the beta "Genetic Communities" feature that AncestryDNA is rolling out may be incorporating this research.

JMcB
02-20-2017, 09:57 PM
..... In my family tree research thus far, I have uncovered one anglicized surname which appears to be more of a humorous failure at English spelling rather than deliberate attempt to look more Anglo. ;) However, there were certainly many instances of surnames changes to avoid discrimination or to assimilate more quickly.

I have one of those in my family. When my Great Great Grandfather arrived in the United States he changed his name from Lo Bianco to White to accomplish both ends. To avoid discrimination and assimilate more quickly.

Calas
02-20-2017, 10:02 PM
I had no idea of the Acadian influence in South Carolina where I still have kin! B)

The Colonies' list of Acadian resettlement http://www.thecajuns.com/exile.htm

I was sent that a couple months ago by a friend who had finally figured out why he had a number of Acadian/French-Canadian/Cajun/Louisianan relatives in his DNA matches when his ancestry was widely central/north eastern States.

AnnieD
02-22-2017, 05:37 AM
The Colonies' list of Acadian resettlement http://www.thecajuns.com/exile.htm

I was sent that a couple months ago by a friend who had finally figured out why he had a number of Acadian/French-Canadian/Cajun/Louisianan relatives in his DNA matches when his ancestry was widely central/north eastern States.

Ah, 280 settled in South Carolina. I always associated the Acadians in N. America with Quebec and Maine and then further south in Cajun-land, or Louisiana. However, this website corroborates the settlement throughout the 13 American colonies:

http://www.emmigration.info/french-immigration-to-america.htm

History of French Immigration to America in the 1700's: The Louisiana Cajuns

"French Immigration to America included a group of people known as the Cajuns of Louisiana. The name Cajun is a corruption of the word Acadian and these French-speaking people who were exiles from the New France colony of Acadia (now the Maritimes). The Expulsion of the Arcadians (1755–1764) occurred during the French and Indian War and was part of the British campaign against New France. The British deported over 11,500 Acadians to the 13 Colonies and some made their way to the French Louisiana colony and established the Cajun culture."

Chronos
03-15-2017, 07:48 PM
Great study.