Origins, admixture and founder lineages in European Roma

[PNGarrison]

European Journal of Human Genetics advance online publication 16 September 2015; doi: 10.1038/ejhg.2015.201

http://www.nature.com/ejhg/journal/vaop/ncurrent/abs/ejhg2015201a.html

Origins, admixture and founder lineages in European Roma
not open access

Begoña Martínez-Cruz1,15,16, Isabel Mendizabal1,15,17, Christine Harmant2,3, Rosario de Pablo4, Mihai Ioana5,6, Dora Angelicheva7, Anastasia Kouvatsi8, Halyna Makukh9, Mihai G Netea10, Horolma Pamjav11, Andrea Zalán11, Ivailo Tournev12,13, Elena Marushiakova14, Vesselin Popov14, Jaume Bertranpetit1, Luba Kalaydjieva7, Lluis Quintana-Murci2,3 and David Comas1 and the Genographic Consortium18

Received 2 February 2015; Revised 20 July 2015; Accepted 11 August 2015
Advance online publication 16 September 2015

Abstract
The Roma, also known as ‘Gypsies’, represent the largest and the most widespread ethnic minority of Europe. There is increasing evidence, based on linguistic, anthropological and genetic data, to suggest that they originated from the Indian subcontinent, with subsequent bottlenecks and undetermined gene flow from/to hosting populations during their diaspora. Further support comes from the presence of Indian uniparentally inherited lineages, such as mitochondrial DNA M and Y-chromosome H haplogroups, in a significant number of Roma individuals. However, the limited resolution of most genetic studies so far, together with the restriction of the samples used, have prevented the detection of other non-Indian founder lineages that might have been present in the proto-Roma population. We performed a high-resolution study of the uniparental genomes of 753 Roma and 984 non-Roma hosting European individuals. Roma groups show lower genetic diversity and high heterogeneity compared with non-Roma samples as a result of lower effective population size and extensive drift, consistent with a series of bottlenecks during their diaspora. We found a set of founder lineages, present in the Roma and virtually absent in the non-Roma, for the maternal (H7, J1b3, J1c1, M18, M35b, M5a1, U3, and X2d) and paternal (I-P259, J-M92, and J-M67) genomes. This lineage classification allows us to identify extensive gene flow from non-Roma to Roma groups, whereas the opposite pattern, although not negligible, is substantially lower (up to 6.3%). Finally, the exact haplotype matching analysis of both uniparental lineages consistently points to a Northwestern origin of the proto-Roma population within the Indian subcontinent.

[T E Peterman]

I hope to see Roma calculators developed that can better ascertain the presence or lack of Roma ancestry. So far, I have used "South Asian" as a proxy for Roma ancestry in Europeans, but I'm not sure how accurate this is. I also have to wonder how much West Asian or Eastern Mediterranean might be present in Roma populations.

Timothy Peterman

[PDHOTLEN]

AncestryDNA gives me "<1%" autosomal South Asian. I've given up trying to figure out where it came from. But I've seen matches going back to Wales and the Chesapeake areas with trace South Asian.

[eastara]

I hope to see Roma calculators developed that can better ascertain the presence or lack of Roma ancestry. So far, I have used "South Asian" as a proxy for Roma ancestry in Europeans, but I'm not sure how accurate this is. I also have to wonder how much West Asian or Eastern Mediterranean might be present in Roma populations.

Timothy Peterman

There has been a study about whole genome Roma ancestry:

Reconstructing Roma History from Genome-Wide Data

http://journals.plos.org/plosone/article?id=10.1371/journal.pone.0058633

South Asian in Europe generally is due to Roma ancestry, but not always. Sometimes East Europeans have Central Asian admixture (like Tartar), which also has a small percentage South Asian.
Sure way to tell a real European Gypsy is to check if he matches other known Roma. They descend from a very small group of people and due to endogamy practically all come as relatives to each other. This is true even if you have a quite distant Roma ancestry the same way as even a small part Ashkenazi gives you a huge number of other Ashkenazi relatives.

[T E Peterman]

My father, his brothers & a couple of their maternal first cousins have South Asian DNA that approaches 3%. I used Lazarus at gedmatch to reconstruct partial genomes of my grandparents and great grandparents. The South Asian portion remained for my great grandmother.

How would one tell, short of asking, if an autosomal match had Roma ancestry?

Timothy Peterman

[eastara]

I share with several full blood Balkan Roma on 23andMe.
There it is possible to see their Ancestry composition and at least some of their relatives with the tool Countries of Ancestry.
If they share multiple segments with other people, they are most probably with Roma origin. One and the same names are repeated with most of the people, who have South Asian as little as 0.5%.
I don't know your background, but some British Roma confirmed by genealogy hardly have any South Asian. 3% is quite high, but as I mentioned it may come not from Roma.
If you have tested with 23andMe, send me a sharing invitation, same nick there, and I can tell you if you match any known Gypsies.

[T E Peterman]

I do have results on 23andme, but none of my other relatives do. I have uploaded autosomal DNA from Family Finder to gedmatch for myself & a number of relatives. My South Asian is merely 0.87% per Eurogenes K13. My father's South Asian is 2.59%. His brothers are 3.03% South Asian & 1.87% South Asian. They have even higher numbers for West Asian & Eastern Mediterranean.

My paternal grandmother came from a family that had slightly darker than normal features for northwest European: coal black hair & brown/ black eyes are common, though not universal. My father has first cousins and second cousins on her side with similarly high South Asian percentages.

My father is 1/4 Swiss through his father's side. The remaining 3/4 is what I call Colonial American, meaning generally English, Scottish, small amounts of Irish, with some Pennsylvania German.

My father's grandfather was Edwin Hall (1862-1950). Edwin had a great-aunt named Rosanna McCormick (d 1889), whose obituary claimed that she was a fortune teller. I don't know if this means that Rosanna was Roma, but I think it raises the possibility. This family lived in Frederick Co., VA & Jefferson Co., WV. There are false traditions on this side of the family of being descended from Charles Wesley & related to Cyrus McCormick.

The Lazarus percentage for Edwin Hall in Eurogenes K13 doesn't report South Asian, but with only three collateral relatives of Edwin tested, his partial genome is a lot less complete than that of his wife.

The Lazarus report for Edwin's wife, Sarah (Eggleson) Hall, showed 2.09% South Asian. She has some relatively untraceable parts of her ancestry in Westmoreland Co., PA.

This family doesn't have any tradition of Roma ancestry.

So, it's all a mystery. The South Asian report is the best explanation that I've ever found for why this side of the family has an intermittent pattern of darker than normal features.

Timothy Peterman

[T E Peterman]

Regarding the sharing invitation, what name should I look for? You can also send an invitation to the following:

Timothy Peterman

[Donald Locke]

I do have results on 23andme, but none of my other relatives do. I have uploaded autosomal DNA from Family Finder to gedmatch for myself & a number of relatives. My South Asian is merely 0.87% per Eurogenes K13. My father's South Asian is 2.59%. His brothers are 3.03% South Asian & 1.87% South Asian. They have even higher numbers for West Asian & Eastern Mediterranean.

My paternal grandmother came from a family that had slightly darker than normal features for northwest European: coal black hair & brown/ black eyes are common, though not universal. My father has first cousins and second cousins on her side with similarly high South Asian percentages.

My father is 1/4 Swiss through his father's side. The remaining 3/4 is what I call Colonial American, meaning generally English, Scottish, small amounts of Irish, with some Pennsylvania German.

My father's grandfather was Edwin Hall (1862-1950). Edwin had a great-aunt named Rosanna McCormick (d 1889), whose obituary claimed that she was a fortune teller. I don't know if this means that Rosanna was Roma, but I think it raises the possibility. This family lived in Frederick Co., VA & Jefferson Co., WV. There are false traditions on this side of the family of being descended from Charles Wesley & related to Cyrus McCormick.

The Lazarus percentage for Edwin Hall in Eurogenes K13 doesn't report South Asian, but with only three collateral relatives of Edwin tested, his partial genome is a lot less complete than that of his wife.

The Lazarus report for Edwin's wife, Sarah (Eggleson) Hall, showed 2.09% South Asian. She has some relatively untraceable parts of her ancestry in Westmoreland Co., PA.

This family doesn't have any tradition of Roma ancestry.

So, it's all a mystery. The South Asian report is the best explanation that I've ever found for why this side of the family has an intermittent pattern of darker than normal features.

Timothy Peterman

Hi Timothy, is your gedmatch number F012148 and F226374 ?
If that is you or one of the kits you are associated with, then that gedmatch number is a match with me.
I am F21343 on Gedmatch.

As I have said to you before, the Lock / Locke paternal lineage is in Y Haplogroup H1a - M82, British Romanichal Gypsy ancestry, and my family certainly resided in Frederick County Virginia and Jefferson County Virginia
( now West Virginia ) who also resided in Berkeley County Va too.

The Lock / Locke family are the only provable Romany family from that region of Virginia and West Virginia to my personal knowledge, but I have long suspected there were other Romany / part Romany individuals besides the Lock / Locke family from that area but I have not been able to prove it yet.

That branch of my Locke family that remained in the Virginia's mid 1700's, still has living descendants living in the Virginia's today. So it is a real possibility your South Asian percentage likely has some blood tie to my Locke family.

Because my Locke family has resided in those counties for over 200 years, I am betting your South Asian percentage is tied to my family some how.

But my Lock family was not alone, there were multiple Romany H1a - M82 male lineages from Colonial Virginia too.

Ingram of Pittsylvania County Va - Halifax County Va are also in H1a - M82.

Bailey of Virginia ( not sure what county ) are also in
H1a - M82.

Campbell of Rockingham County Va are also in H1a - M82

just to name a few from Virginia in H1a - M82.

You got some great clues of provable Romany family's residing in Virginia / West Virginia, any number of those family's could easily explain why participants with Virginia ties are finding South Asian percentages in their autosomal reports.

[Donald Locke]

That is the one thing I keep trying to express to everyone, to look for Romany / Romany mixed participants in your FF matches because those matches are the Romany connection you should be focusing on.

South Asian percentages by its self doesn't prove a Romany connection in your tree, but having a match to a Romany or Romany mixed person in your matches is exactly whom you should be focused on in your genealogical paper trail research.

While I am not a FF match to Timothy according to FTDNA FF matches, we certainly do show up as a match on Gedmatch, and because my Locke family has resided in Frederick County Va, Berkeley County Va, Jefferson County Va, Clarke County Va for over 200 years, anyone matching me really should be keeping an eye out for a Lock / Locke family tree connection because we are the only provable Y Haplogroup H1a - M82 paternal lineage residing in those counties of Virginia.

If you take the time to read the professional Romany DNA studies, they all say the exact same thing, Y Haplo H1a - M82 is considered the "founder" population, so when you have an autosomal match with a provable H1a - M82 participant, that is the lineage your paper trail research should focus on because that lineage is the most likely connection in your tree.

[Donald Locke]

To date every Y Haplogroup H1a* - M82 male lineage identified from the Colonial American era, are of the Romanichal clan of England and Scotland.

Yes there were many Romanies whom migrated later during the 1850's - 1900's Romany migration whom came from all across Europe to the USA. But as far as the Colonial American Romany, every lineage I personally know about and have researched are from the Romanichal clan of England and Scotland which is a direct result of the anti Gypsy law era of England and Scotland, both of which were forcefully transporting Romanies as convicts to the American Colonies as early as the 1670's - 1700's era.

Those autosomal DNA participants who find some South Asian percentage, really should be looking at those provable Romany paternal lineages in H1a - M82 because some where in any one of those family trees could very well be the connection in your tree to help you prove that Romany connection.

[georgian1950]

I am F21343 on Gedmatch.

Donald, do you know your mtDNA haplogroup?

Thanks,

Jack Wyatt

[Donald Locke]

georgian1950
Jack, no I do not know my mt Haplogroup yet, but my mother is of Germanic origins so her Haplo wouldn't do me much good as far as the Romany connection.

[georgian1950]

georgian1950
Jack, no I do not know my mt Haplogroup yet, but my mother is of Germanic origins so her Haplo wouldn't do me much good as far as the Romany connection.

Thanks, something other than the Romany connection has caught my interest. The mtDNA haplogroup might have been some help.

Jack

[Donald Locke]

@ Timothy

2 Lock brothers remained in the Virginia's.
George Lock(e) whom married Sarah Hiatt
John Lock(e) whom married Esther French

George and John are as far as I know the only branches of my tree to have remained in the Virginia's and there are many living descendants of those 2 Lock men living in those states today.

The rest of the Lock family moved to Mercer County Kentucky by the 1780's and settled in Barren County Kentucky by 1799.

Some of the brothers settled down in Jackson County Tennessee, Wayne County Illinois, Barry County Missouri.

11 Locke men in the USA, descendants of Richard Lock of Frederick County Virginia have Y DNA tested, all 11 Locke men proven to be in Y Haplogroup H1a - M82.
So anyone with family ties to Frederick County Va, Berkeley County Va, Jefferson County Va, Clarke County Va., Barren County Ky, Hart County Ky, Jefferson County Ky, Jackson County Tn, Wayne County Illinois, Barry County Mo, Polk County Mo. with South Asian percentages really should be looking for a Lock / Locke family tree connection because proving that paper trail connection could easily explain your South Asian percentage.

But again, there were multiple Y Haplo H1a - M82 paternal lineages from the Colonial era from Virginia and South Carolina, any number of those trees could also help explain your South Asian percentages.

https://www.familytreedna.com/groups/locke-dna-project/about/background

Group 2 in the Locke project is my family, and 3 of those Locke participants are direct descendants of George Lock and Sarah Hiatt of Frederick County Va, Berkley County Va, Clarke County Va, Jefferson County Va.