Search This Blog

Loading...

Monday, May 7, 2012

Eurogenes' GEDmatch ADMIXTURE ancestry test guide


Update 27/11/2013: I've made the new K13 the default Eurogenes admix test at GEDmatch. It seems to hit the spot for most people. See here.

Update 07/10/2013: An upgraded version of the popular EUtest is now available at GEDmatch. See here.

Update 11/10/2012: The Jtest and EUtest are now available at GEDmatch, and these have superseded most of my tests offered there, except those that include Amerindian references samples. Also, the Jtest and EUtest come with a variety of "Oracle" ethnic matching tools. For more info see here.

...

First of all, below is a list of the geographic locations or ethnic groups in which the clusters from the Eurogenes' Gedmatch tests peak. I can't post any population averages or maps at present, but will do so as soon as possible. I really need more people to fill in their ancestry details at Gedmatch when they run these tests, otherwise producing more detailed guides won't be possible.


K9
South Asian - South India
Caucasus - Georgia
Southwest Asian - Bedouin
North Amerindian + Arctic - Northwest America
Siberian - Central Siberia
Mediterranean - Sardinia
East Asian - Eastern China
West African - Nigeria
North European - Lithuania


K10
South Asian - South India
Caucasus - Georgia
Southwest Asian - Bedouin
North Amerindian + Arctic - Northwest America
Siberian - Central Siberia
Mediterranean - Sardinia
East Asian - Eastern China
West African - Nigeria
East European - Belarus
North Atlantic - Ireland


K11
South Asian - South India
Caucasus - Georgia
Southwest Asian - Bedouin
North Amerindian + Arctic - Northwest America
Siberian - Central Siberia
Mediterranean - Sardinia
East Asian - Eastern China
West African - Nigeria
Volga-Ural - Western Volga
South Baltic - Lithuania
North Atlantic - Ireland


K12
South Asian - South India
Caucasus - Georgia
Southwest Asian - Bedouin
North Amerindian + Arctic - Northwest America
Siberian - Central Siberia
Mediterranean - Sardinia
East Asian - Eastern China
West African - Nigeria
Volga-Ural - Western Volga
South Baltic - Lithuania
Western European - Western Ireland
North Sea - Southern Norway


K12b

Western European - Cornwall
Siberian - Central Siberia
East African - Masaai
West Central Asian - Balochistan
South Asian - South India
West African - Nigeria
Caucasus - Georgia
Finnish - Eastern Finland
Mediterranean - Sardinia
Southwest Asian - Bedouin
North European - Lithuania
East Asian - Eastern China


K13

North European - Lithuania
West African - Nigeria
Mediterranean - Sardinia
Northeast African - Ethiopia Oromo
North Eurasian - Central Siberia
South Asian - South India
Southwest Asian - Bedouin
Pygmy - Mbuti Pygmy
Caucasus - Georgia
East Siberian - Koryaks
East Asian - Eastern China
Amerindian - South America
West Central Asian - Balochistan


Hunter-Gatherer vs. Farmer (Gedmatch edition)

Anatolian Farmer - Western Caucasus
Baltic Hunter Gatherer - Lithuania
Middle Eastern Herder - Bedouin
East Asian Farmer - Eastern China
South American Hunter Gatherer - South America
South Asian Hunter Gatherer - South India
North Eurasian Hunter Gatherer - Central Siberia
East African Pastoralist - Masaai
Oceanian Hunter Gatherer - Papua New Guinea
Mediterranean Farmer - Sardinia
Pygmy Hunter Gatherer - Mbuti Pygmy
Bantu Farmer - West Africa

Now, the K12 test is the most useful analysis in the context of my project, which focuses on the genetic substructures of groups from north of the Alps and Carpathians. Therefore, most of this post will be dedicated to that test, and especially to the European and West Eurasian clusters it features. Indeed, below is an MDS plot showing synthetic samples made from the allele frequencies of the seven West Eurasian clusters from the K12.




It's pretty easy to see what's going on there. The four North European clusters are positioned close together on the left of the plot. However, the Western European and North Sea clusters both show significant affinity to the Mediterranean cluster, by pulling towards its direction.

Note also that the Mediterranean and Caucasus clusters are placed about half way between the South Baltic and Southwest Asian clusters across dimension one. This is in line with geography, because the Mediterranean Basin and Caucasus Mountains are located between Northern Europe and the Middle East.

However, the Mediterranean and Caucasus clusters are sharply differentiated across dimension two. This also fits geography, because there's obviously a huge distance between the populations of the Western Mediterranean (in which the Mediterranean cluster peaks) and the Caucasus.

Below are a few spatial maps, showing the distribution of selected clusters from the K12 analysis (courtesy of Eurogenes project member FR7). The data is still fairly limited, so those of you who want your countries/ethnic groups represented, please encourage more people to take the Eurogenes K12 test at Gedmatch, and state their ancestry when they do.


158 comments:

jackson_montgomery_devoni said...

Sorry to repeat myself David but I will not ask this question again. This will be my last time asking it for these runs. I am an admixed European individual. I am 25% Italian, 25% Finnish and 50% Irish/British by known paper trail ancestry. Which tests/analysis are meaningful for an individual such as myself?

Davidski said...

^ Considering your mixed ancestry, and the specifics of that, I think the best tests for you would be the K9, K10, K12b_new and K13.

The other runs are only really useful for you when looking at segments, by cross checking all the results, and seeing what patterns emerge.

jackson_montgomery_devoni said...

Okay once again thank you very much David!

~Elizabeth~ said...

Hello Davidski,
Thank you for the "list of the geographic locations or ethnic groups in which the clusters from the Eurogenes' Gedmatch tests peak".

:)

teri said...

Hello,
If you have time, I have two questions for you - 1. If the Mediterranean portion is always represented by Sardinians, why does the percentage change for me with the various admixes? 2. If I know from the paper trail that I have colonial American of heavy English ancestry, plus some known Irish, German and Jamaican Euro/Afro mix, could you advise me the best admix? I was thinking maybe K12 or K12B...
Thanks for making this available!

Davidski said...

^ The Med cluster always peaks in Sardinians, but not always at the same levels. That's because when the results break up into more Ks, some of the Med alleles go to the new clusters, like the Western European or the North Atlantic.

People who are mixed are better off using the results from the lower Ks. At higher Ks, many of the clusters are differentiated by very low genetic distances, so if you're really mixed, your results will look like soup.

Unknown said...

Hello,

First of all, thank you very much for all your hard work, and for making these tests available on GEDmatch.

I am also very mixed: 50% Irish, 25% German, 12.5% French, and 12.5% unknown. I have been researching my family history via paper genealogy for years and have 7/8 of my ancestry pretty thoroughly documented, but one of my great-grandparents has been a complete brick wall. One of my main motivations for taking the DNA test was to attempt to get an idea of what kind of ancestry that unknown 1/8 might be. Do you have any advice for me?

Thank you.

Davidski said...

^ The best course of action would be to find someone who is 50% Irish, 25% German and 25% French, or close to that, and compare your results with theirs from many different tests (basically all the admixture tests at Gedmatch).

If there's a clear and consistent difference in the results, then you can then try and work out which way you're being pulled away from that 50% Irish, 25% German and 25% French mix. That might give you some clues about that 12.5%, especially if you also see signs of the same trend in the Ancestry Finder at 23andMe.

So, for example, you find that across all the admixture tests, you show clearly higher South Baltic or East European scores. Then you check your Ancestry Finder results, and you see that you share some fairly large segments with people from the Baltic states. Then you have a look at what the chromosome paintings are showing in those areas where you share the segments, and they show a lot of South Baltic and similar clusters. All of that would indicate that the 12.5% came from somewhere around Lithuania.

Unknown said...

Thank you very much for your reply.

Very interesting. If I'm understanding you correctly, the hard part will be finding other people with that 50/25/12 mixture who I'm not related to. I'm not sure how to go about doing that, other than putting out a general request on something like the FTDNA forums (my test was with FTDNA). The more people I can find with that mix, the better, correct?

FWIW, I've thought about testing with 23andme, or Ancestry when that test becomes available, to see if that could shed any new light on my case. I also have five full siblings and my understanding is that the more of us who get tested, the more "complete" picture of our ancestry will emerge.

Thank you again.

Unknown said...

Hi David,
Really glad the averages have arrived, it' really useful. I am UK42 and i was investigating this as well as chromosome paintings on GEDmatch, and i was wondering if you would consider that an excess of Southwest Asian, Mediterranean and African components but an average component of Caucasus (this appears consistently) might indicate a North African origin for my unknown ancestry as opposed to plain Middle-Eastern. It is quite convincing especially as i have recently come across a significant chunk of African at one end of Chromosome 13.

Basically to cut it short, would you consider this apparent absence of Caucasus which is so significant in Middle-eastern populations, as well as the significant presence of an African component to be indicative of North African ancestry as opposed to Middle-eastern?

I've found your K=12 to be very useful for all of my ancestry, many of the others are useful for particular parts but this one seems to cover it all!

Kind Regards,
Sam Jackson

truth said...

No spanish ? Is strange, because I know at least one spaniards who typed his ancestry on Gedmatch.

Davidski said...

^ Sam, yes, if you have more of the components that you mentioned, and less Caucasus, then this indicates more influence from the south and southwest, rather than the southeast.

I suppose the next step would be to check your Ancestry Finder results to see if you have any interesting hits in areas of your genome where these exotic components dominate. You might find some Sephardi Jews there, or maybe someone from Malta?

Unknown said...

Hi David,
Thanks for the clarification - I'll look a bit deeper into my matches a bit later on, although i do have a couple of small segments matched with people in Spain, although just recently i got 0.2% show up as a match with someone (or perhaps more than one?) with all 4 grandparents from Haiti, quite interesting!

Malta is an interesting idea - Hadn't thought of that and i'll give it a look too, Sephardic Jewish is my explanation up until now so in terms of genaeology it would be nice to make a connection, although i'm aware that whatever it is, it's unlikely to be as clear cut if they weren't registered as Jewish, and if it is North African, well that will be difficult...

Cheers!

Amanda S said...

Why would you characterise your Oceanian as a hunter gatherer? Although there are some small hunter gatherer groups in Papua New Guinea the majority of the population have been farmers for thousands of years.

Davidski said...

^ Europeans have been farmers for thousands of years too, save the Saami of Northern Scandinavia. But it'd be kind of redundant to classify groups as farmers in this test, based on the fact that they've had farmer ancestors since the late Neolithic at least, because that wouldn't leave any hunter-gatherers.

Amanda S said...

Yes, but New Guinea seems to be the third oldest area for the development of agriculture after the Fertile Crescent and the Yangtse and Yellow River basins. Crops domesticated in New Guinea such as yams, taro, sugarcane and bananas were spread around the Indian Ocean to reach India and Africa.

daspit said...

I am curious where Greece and Turkey fit into the K12 descriptions above. Would either of them be Mediterranean, Caucasus, or southwestern Asian?

giorgoa said...

I am from greece and i wonder if it is possible that some of my ancestry that appears to be western european and north sea could actually mean that i have more mediterranean and south baltic admixture.

Davidski said...

^ Well, there's really no such thing as Mediterranean, South Baltic or North Sea, etc.

These are just names for allele frequencies that have no borders.

The only way to check whether you show influence from the south, west, east or north is to compare your scores to other people from similar parts of Greece.

Jaime Rendon Hernandez said...

Per FTDNA I am of mixed, Native American, African, Finnish, Russian and Jewish Palestinian ancestry, what tests would accurately work for me David?

Thank you,

Jim

Ronda Miller said...

Per FTDNA & a fairly good paper trail I am roughly 25% Mayan with the rest being Finnish, Russian, Spanish, Tuscan. Which of your tools would best for me?

Thank you,

Ronda K Miller - Lead South Dakota

S Mark Roberts said...

Hi David,

Steven Mark Roberts here; kit M083210. I am a 'newbie' to gedmatch.

I have been tracing my family tree for over twenty years, back when I had to go to the library to look at microfiche.

Except for one German gg grandfather, all of my family has been here since before the Revolution. And except for a NY Dutch ancestor, ALL of my ancestors were from the British Isles; primarily English, Scottish and Welsh. I am the classic 'WASP.'

There is a family legend of Cherokee Indian.

I first did DNATribes, but I found their information to be useless. ( I would be glad to share my results with you for your research)

I then did FamilyTreeDNA; it showed that I was 94% Oracadian and 6% Middle Eastern, which I suspect may be Jewish. I suspected this came through my German gg grandfather, as most all other ancestors were British, and the British expelled Jews, and by the time Jews were back in Britain, my family was already in the US.

To test this theory, I tested my gg aunt's DNA, and she has around the same amount of middle eastern.

I did ancestryDNA, and it showed 60% viking, 20% central europe and 19% British Isles.

I did 23andme, and it showed 99.5% European and .5% american Indian.

I have uploaded my 23andme raw file; I will upload my raw file from familytreeDNA when they allow it (it is 'broken' as of 022813).

Since I am 99.5 percent european per 23andme, and 95% orcadian per familytreedna and 98% european (2% unknown) per ancestry.com, which test of your admixture should I use?

If you'd like to see the 'reports' that DNATribes sent me, just email me at mrobertsbham2000@yahoo.com and I will be glad to send them to you.

thanks so much.

Mark Roberts

Davidski said...

^ I just had a look at your Eurogenes results at GEDmatch.

You basically come out like someone of mostly English descent, but with considerable and probably recent ancestry from the Northwestern European mainland, like from Denmark, Frisia, or perhaps even Holland north of the Rhine or Northwest Germany.

You're definitely not typically Orcadian, and you don't have any recent Middle Eastern ancestry. The reason you're coming out Orcadian/Middle Eastern at FTDNA is because of the way their (rather chunky) algorithm works. Basically, it attempts to place you along the genetic cline that runs between Northwestern Europe and the Middle East, and its saying that you're a few per cent more Middle Eastern (or rather Neolithic) than the average Orcadian. I'd also add that AncestryDNA seems to be overestimating your "Viking" or rather Scandinavian roots by at least 10%.

The Jtest Oracles dont pick up any clear Ashkenazi admixture, but the chromosome painting suggests there might be a couple of minor segments from a distant Ashkenazi ancestor, like on the left tip of chromosome 6, and perhaps on the left tip of chromosome 19. If these are genuine, then you should see that confirmed at 23andMe using the Ancestry Finder and the Ancestry Composition. In other words, if the signals are real then those tools will show Ashkenazi hits in the same areas as the Jtest chromosome painting does.

There's probably a stronger case for minor Amerindian admixture. It's tiny overall but shows up consistently, and there seems to be a large Amerindian segment on chromosome 13. Keep in mind that Europeans, and especially Northern, Central and Eastern Europeans, apparently share prehistoric ancestry with Amerindians, so it's always possible that this is a signal of that ancient relationship. However, I'd be surprised if the Amerindian segment on chromosome 13 wasn't due to recent Native American ancestry, and I'm guessing that it shows up in the Ancestry Composition at 23andMe.

Richard Robbins said...

@Davidski

I am new to all this DNA testing, and I've been having fun running the various tests that are available on Gedmatch. I am interested in knowing which test is more useful with my admixture? I am getting different reports from various tests, and I'd like to know which one is best for accurate results? My ancestry is Polish on my mom's side, back 5 generations that I know of, but only her father's tree. My father's side comes from Hungary or Croatia, with Austrian and German mix (his mother's side). Father's side is Colonial american, with ancestry from Scotland and Ireland. Kit is M111825

Davidski said...

The best tests for Europeans are the Jtest and EUtest because the Oracles linked to those work correctly for everyone (not just those who are members of various projects). The reason for this is that these tests don't suffer from the calculator effect.

However, if you're partly Native American, then this will be substituted in the Jtest and EUtest with Siberian admixture and a Siberian ethnic group in the Oracle results. That's because there are no Amerindian reference samples in these tests.

Angela Fontes said...

Hi David :)

Angela here. Gedmatch number: M094348 I am very new to all of this. I have done my geneology and most ancestors are from England and some from France and others for Ireland. I can't trace them all. I also have a 3x great grandmother that I have a picture of that was born in 1844 in San Francisco. She looks Native American. Her husband was born in Mexico but he was white looking so I guess he might not have mixed with the natives too much. I have been using Gedmatch... Very confused how to read the information I see. I do see Native American DNA and I also see some African DNA. I am curious about that. I am also confused by my European DNA. I have a lot of mediterranean and I don't know where that is from. I was wondering if you could take a lot at my Gedmatch number and tell me what you know about what you see. Thank you for any insite you can provide :)

Robert Shawn Rowland said...

Hi David,

First off...thanks for all your hard work. It is something that has come to help me out greatly lately. I don't really understand much of this so I'll just tell you what confuses me. I took the Ancestry.com test and came up with 98% Scandinavian. I know they have a big Scandinavian problem anyway so there is that. Then I took the raw data and put it into several of the tests last night to see what would come up and all of them are very similar...but I'm not sure what it means. Here are a couple of the tests:
K 12
Population
South Asian 1.48%
Caucasus 5.63%
Southwest Asian 0.72%
North Amerindian + Arctic 0.60%
Siberian -
Mediterranean 13.99%
East Asian -
West African 0.52%
Volga-Ural 2.90%
South Baltic 11.05%
Western European 31.02%
North Sea 32.09%

K 12 b
Population
Western European 46.30%
Siberian 0.51%
East African -
West Central Asian 0.80%
South Asian 2.17%
West African 0.65%
Caucasus 6.98%
Finnish 4.92%
Mediterranean 15.98%
Southwest Asian 1.39%
North European 20.30%
East Asian -

K 36
Population
Amerindian -
Arabian -
Armenian -
Basque 0.50%
Central_African -
Central_Euro 5.56%
East_African -
East_Asian -
East_Balkan 2.14%
East_Central_Asian -
East_Central_Euro 5.44%
East_Med -
Eastern_Euro 3.14%
Fennoscandian 6.83%
French 8.55%
Iberian 11.79%
Indo-Chinese -
Italian 12.79%
Malayan -
Near_Eastern -
North_African 0.52%
North_Atlantic 19.31%
North_Caucasian 1.53%
North_Sea 17.87%
Northeast_African -
Oceanian -
Omotic -
Pygmy 0.53%
Siberian -
South_Asian 0.70%
South_Central_Asian -
South_Chinese -
Volga-Ural -
West_African -
West_Caucasian 0.69%
West_Med 2.11%

Ok, so what am I looking at? I'm especially unsure about the Italian and Iberian percentages. I know that can refer to deep ancestry, especially for someone with a lot of English ancestry...but they seem high...are they?

Davidski said...

Well, these results are meaningless unless you...

a) cross check the component levels with lots of other users to try and find patterns and anomalies.

b) look at the chromosome paintings and cross check them with half-segment matches with other users to perhaps find local areas of your genome that carry admixtures from specific areas of the world.

The other thing you should do is run the EUtest and Jtest and study your Oracle results. This will help to put your overall ancestry into perspective in the context of the Old World.

ApeQuake said...

My son is 1/2 Thai (his mom - all 4 grandparents born in Thailand) and 1/2 Euro Mutt (me). Here is his admixture:

Indo-Chinese 15.98%
Malayan 14.68%
Central_Euro 7.82%
South_Chinese 6.65%
North_Atlantic 6.11%
North_Sea 5.63%
East_Asian 5.35%
Italian 5.14%
Fennoscandian 4.99%
East_Central_Euro 4.32%
South_Asian 3.99%
Iberian 3.28%
Siberian 3.24%
French 3.13%
Basque 2.81%
East_Balkan 2.70%
West_Caucasian 2.19%
East_Med 0.84%
Armenian 0.54%
Pygmy 0.52%
Oceanian 0.09%

Erik Schaefer said...

David,

I've created a Facebook group for people who are new to GEDMatch and just learning how to use it. Many of us are facing a steep learning curve with the Admixtures and knowing what all of the different options mean. Would you mind joining our group and giving an overview of all your different test options for Eurogenes? Link to our group:

https://www.facebook.com/groups/405531739562579/

Davidski said...

Unfortunately, I don't have a Facebook account. The blog entries and comments I've made here should cover all the basics.

But the best way to get grasp how to work the tools and interpret the output correctly is to compare notes and learn from each other.

It might seem like a steep learning curve now, but with a bit of experience everything will eventually make sense.

Nicholas Brady said...

Hey David, my GEDmatch kit # is A238229. On a few of the tests, I'm getting a small percentage of North Amerindian + Arctic (around 0.4%). As far as I'm aware, I don't have any Native American ancestry in my tree, though I am descended from original colonists so it could be there somewhere. Could this be correct? The rest of the results I understand, seeing as I have ancestry from Ireland, Scotland, Ukraine, Slovakia, Italy, Germany, etc.

Davidski said...

I'd say the 0.4% looks like noise. Keep in mind that Europeans share deep ancestry with Siberians and Amerindians, so the algorithm can't always infer correctly the origins of some of the alleles. They might well be from an ancestral third party that no longer exists.

CatchingAllTheSouls said...
This comment has been removed by the author.
Unknown said...

Hello Davidski and thank you so much for all your hard work, that allows users like me to have the opportunity to use.

A person who gives lots of advice to others on the subject of DNA, said this to me: "admixtures, ethnicity, or more technically, bio-geographical ancestry (BGA) analysis, is little more than novelty at this point for most Americans." And this person is not the only person to allude to admixtures as being a 'novelty' or 'entertainment'. I'm not understand why people like the aforementioned are advising people this way. I'm really new to all this DNA stuff and find these type of responses really off-putting. I don't believe that knowledgeable people like yourself, have spent countless hours of time and effort to create DNA admixture analysis programs for just 'novelty' & 'entertainment'.
Firstly I'm not 'American'...I'm Brit born as well as all of my maternal family.
I also submitted my Ancestry raw data to Dr. McDonald, who I do have great respect for. This is what he wrote to me: "Most likely fit is 100% English (Western Europe)which is 100% total Europe.
The location error = 0.005618 with 1 group
The following are possible populations, most likely at the top: English= 1.000 but it is just a little bit more Continental European flavor than the average English."
Though his plot graphs tell a bigger story about my DNA.

These are my results using your EUtest Admixture:
Population
SOUTH_BALTIC 11.54%
EAST_EURO 10.40%
NORTH-CENTRAL_EURO 26.19%
ATLANTIC 27.89%
WEST_MED 13.68%
EAST_MED 6.90%
WEST_ASIAN 3.21%
MIDDLE_EASTERN -
SOUTH_ASIAN -
EAST_AFRICAN 0.19%
EAST_ASIAN -
SIBERIAN -
WEST_AFRICAN -

To add, my maternal grandfather (born in England) thru his father (and beyond)are Romany gypsy. Also, my Brit biological father is unknown to me. Davidski, could you please help me to understand/clarify what I see (results) and what I am hearing from others? Thank you for taking the time to read this.

Brooke Fiechter said...

David, thank you for your assistance with our genealogy research! I don't know what I'm doing.... my results from K9 are: Caucasus 9.09%, Southwest Asian 2.57%, Siberian 0.35%, Mediterranean 29.50% and North European 58.49%.
My great-great-great grandmother was "supposedly" Cherokee. Could the Siberian be representing that or is it most likely a family myth? Also, is there any way of determining a greater population specificity for all of my populations? For example, many different countries comprise Caucasus. I would like to know what one exactly. And do several countries comprise Mediterranean? I am confused because somebody said something above that made it look like just Sardinia does.
Thank you again!

Brooke Fiechter said...

OK, so I ran the K36 test and got a VERY specific idea, as there were 14 different entities listed! However, is that the best test for me? I am confused at why there are so many and what is the most accurate one for me. Help!

Davidski said...

I think it's best to start off with the EUtest oracles. These deal directly with what most users are interested in, which is what their DNA says about their ethnic origins. Just keep in mind though, that the results are limited by the reference samples available, so sometimes you'll need to think whether a result might be a stand in for something else that's not there. For example, you might know that your grandfather was Welsh, but the oracles are showing a lot of Cornish in the results. But of course there's no Welsh sample, and Cornwall isn't far from Wales, so in this context Cornish is Welsh.

After that, it might be useful to go thorough some of posts and comments here, at your own pace, to get an idea what all of these admixture results represent, and the various contexts in which they should be viewed in. The basic thing to understand is that the raw admixture results represent deep ancestry, often going back to the Neolithic, and the orcales then try to predict what that means in the context of modern ethnic groups.

Brooke Fiechter said...

Thank you so much, David! So are the percentages I see representative of my actual percentage as it is today? For example, let's say someone had a result of 10% French.... is it accurate to say this person is 10% French in terms of their ancestry? Also, my Jtest showed I am 2 point something % Ashkenazi; you said somewhere, though, that at that percentage it is probably an error, didn't you?

Davidski said...

Getting 10% French in the admixture run doesn't necessarily mean 10% French ancestry. It just means that 10% of the genome is very similar to the French reference sample.

To see if you have recent French or Ashkenazi ancestry, run the EUtest and Jtest and check the Oracle results. If French and Ashkenazi show up consistently in the Oracle results, then recent ancestry from these ethnic groups is likely.

Brooke Fiechter said...

Thank you, again! By chance do you have a Swiss reference population? I know that a significant part of my ancestry is Swiss, but I have absolutely nothing in these results reflecting that. And regarding what you said about 10% French meaning that 10% of the genome is very similar to the French reference sample.... if hypothetically I am NOT part French, what is the alternative explanation for that similarity then?

Brooke Fiechter said...

Also, if you choose to look up my data, it is kit # F279062

CatchingAllTheSouls said...

Why does K12 show an even mixture of Western and Northern European while other calculators show different mixes? Honestly K12 seems more accurate given what I know about my ancestry vs EUtest which blows some things way out of proportion, especially when I look at the Oracle results. I am American so that could be why but.. yeah.

CatchingAllTheSouls said...

And Brooke we have similar K12 results. I just ran your kit number through. I have high North and Western European results by about 1 percent or 2.. doubt that it matters but you are 1 percent higher in Volga Ural. I am not even sure if this implies that we have common ancestry or not considering that these regions are quite vague and do not specify exact cultures or nations.

Davidski said...

Many of the tests carry the same names for their components, but these components are never identical in terms of allele frequencies.

That's why the component names should never be taken too seriously, and only really used as a rough guide.

auntsha said...

My maternal ethnicity is 27% Irish/ 18%Acadian/ 5% Spanish...My paternal is Unknown except for all the test indicating half Southern Eur. My Gedmatch is #M161101 and seems to indicate on the tests N Italian. There is a Serbian/Croation element with Tribes and one of McDonalds analysis which muddles the water for me seeing clearly of just what ethnicity is real. My AF is heavily Acadian because of the intermarriage thing. Italy does come out ahead with 6/995 matches with 4 grandparents from Italy. I would appreciate your perspective on my results. What is the best test for me when a person does have an unknown factor? Thank you.

Ryan said...

Hello David,

My kit number is M201319, I have never met my father, so I developed an interest in genetic ancestry to try and find out a little about where my ancestors may have came from. I have tried both the EUtest and the Jtest, but am having a little trouble interpreting what the results mean. My results are:

SOUTH_BALTIC 12.47%
EAST_EURO 11.03%
NORTH-CENTRAL_EURO 25.07%
ATLANTIC 27.82%
WEST_MED 13.36%
EAST_MED 2.56%
WEST_ASIAN 5.20%
MIDDLE_EASTERN 0.13%
SOUTH_ASIAN 0.87%
EAST_AFRICAN 1.46%

My question is, does the various matches from Asia, Africa, and the Middle East point to the Moorish invasion of ancient Iberia, or is it actual ancestry? Thanks so much for you time, I hope to hear from you soon.

Ryan

Davidski said...

Have you tried the Oracles linked to the EUtest and Jtest? They should give you some clues about your ethnic origins, as opposed to ancient admixture components that make up your genome.

Keep in mind, however, that many ethnic groups in Europe are very similar genetically and overlap quite a bit.

Ryan said...

Thanks for the quick response. I am looking at the Oracle results right now. I can't quite figure out what a lot of the abbreviations mean. For example what is IE? Is there a website that says what countries or regions these abbreviations stand for?

Davidski said...

IE is Ireland.

http://www.uspto.gov/patft/help/helpctry.htm

Ryan said...

Sweet. Thanks so much, this should keep me busy for a while. Thanks for your help.

Squirrel said...

Hi David,
Can you explain why every admixture calculator I've tried (including yours) shows small amounts of Native American but 23andme shows me at 0%?

Davidski said...

There are two possible reasons:

1) 23andMe relies on relatively precise haplotype matches to confirm admixture, but ideally this approach requires lots of samples (hundreds or even thousands) to capture all the haplotypic diversity in the ancestral populations chosen, and 23andMe only has 108 Amerindian samples, including none from north of Mexico.

2) Admixture tests like those at GEDmatch rely on allele frequencies to analyze ancestry, which is basically a method which in most cases pushes back the timeframes of admixture being looked at by thousands of years. What this means is that you're likely to see signals of very ancient admixtures or common ancestry from prehistoric ancestral groups which no longer exist. That's why in tests like these it's crucial to compare your results to those of others to get an idea of how to spot recent admixtures.

Anonymous said...

Hi David,
I was wondering if you had the time, if you wouldn't mind looking at my kit number, and telling me if the small amounts of native american ancestry, that the admixtures are picking up, are worth noting. I have documented and proven native american heritage, from both sides of my family, but for some reason Ancestry.com DNA test, picked up none. Unless of course it lies in the "uncertain" percentage. Ancestry.com has me as 40% central european, 34% british isles, 14% eastern european, 10% scandinavian, and 2% uncertain. Upon talking to others, I have heard that ancestry's admix tools are horrible...LOL i'm not sure how true that is, but I was hoping that you could maybe help me determine whether or not I truly do have any native american ancestry. My kit number is A471422. Thank you so much!

Kayla

Davidski said...

Yes, it seems you do have some Amerindian ancestry, and I'd say it looks real, but it's just above the noise margin in these tests (which is around 1% for continental ancestry).

It appears to be confined to small segments on chromosomes 1, 8, and 13, and perhaps also 20 and 22.

I suspect that you if got a test done at 23andMe you'd probably get around 0.5% Amerindian on the Ancestry Composition, with the segments on chromosomes 1, 8 and 13 being the most likely to be flagged.

CatchingAllTheSouls said...

I have a MUCH higher Western European average than all of the populations. My family is reportedly English and Doug McDonald concurs. Now McDonald's analysis also points to a possible French admixture of which would consist of about 30 percent of my ancestry if that admixture was correct. The top admixture was 100 percent English. Now I have no known Cornish or Welsh ancestors. In fact all of the surnames in my family are clustered in northern and eastern England. However, two surnames in my family break that trend by one being common in northwestern England and the other being most common in Ireland. Is it possible that I may have heavy Cumbric lines? North and northwestern England were Cumbric-British lands prior to the Saxon takeover of Britain.


Western European 43.53%
Siberian 0.44%
East African -
West Central Asian 0.73%
South Asian 1.68%
West African 0.22%
Caucasus 5.87%
Finnish 5.56%
Mediterranean 17.99%
Southwest Asian 1.66%
North European 22.31%
East Asian -

CatchingAllTheSouls said...

Ah well I was looking at K12 averages which explains that.. here is my K12 which still looks a bit odd. I am of British descent. Why such a high Mediterranean percentage?


Population
South Asian 1.41%
Caucasus 4.94%
Southwest Asian 1.08%
North Amerindian + Arctic 0.11%
Siberian -
Mediterranean 15.65%
East Asian -
West African -
Volga-Ural 5.05%
South Baltic 12.93%
Western European 29.81%
North Sea 29.02%

Davidski said...

Druids?

http://www.bbc.co.uk/news/science-environment-11421593

http://news.nationalgeographic.com.au/news/2010/10/101013-stonehenge-burials-boy-science-mediterranean/

Scott: said...

I have noticed that the Native American populations in the admixture tests both here and for the others available at www.gedmatch.com do not seem to have reference populations for Eastern Native Americans, only Northwestern, Western, Southern, and on down in through the Central Americas into South America.

Is this just due to lack of decent samples?

I ask because I consistently show up with 1% to 2% Native American regardless of admixture test (as long as they have a Native American reference population in the testing), but the family stories and lineage would require that part of my ancestry to one of the Eastern Native American bands.

Thanks,

Scott

Davidski said...

There's a lack of Amerindian samples from the US and Canada because the native communities there don't want to be sampled. But considering that there isn't a huge genetic difference between unadmixed native samples from, say, East Greenland and Peru, then it's unlikely that Amerindian admixture from the eastern US would go unnoticed with the currently available reference samples from Central and South America. Also, the admixture that doesn't show up accurately as Amerindian is usually classified as Siberian or East Asian, rather than any type of European, and that's usually a small part of the total anyway.

matt c said...

I took the test as well and from what i know about my ancestry, my dad was romanian and russian jewish, my mother german and french, we live in the united states though my dad had family in canada for a while. We show predominantly northern european followed by med, also showing as admixture of artic amerindian and siberian? Is this correct? Some tests show I have asian and actually underreport any middle eastern ancestry with most places estimating 11-29 percent, 23and me says I have no middle eastern ancestry and reports me as being only european and ashkenazi (43 percent) with about 0.7 percent balkans and 2 percent eastern european. My dads family has black hair, olive skin and looks like they came out of the middle east or greece or spain. The differences with the tests is kind of confusing.

Anonymous said...

thank you lots!!!

Natalie Baker said...

Hi David,

My gedmatch kit # is m503215 and I am trying to determine possible E. European ancestry and the source of supposed 1.3 % Sub-Saharan African ancestry. I am interested in your interpretation -EU test shows lots of E. European and then verifies the African, but I am not sure of what to make of the E. European and if the SSA is from US slave origins given my known paper trail?

My paper trail shows me as about 21% English/Irish, some Dutch and German, and then French and Cajun French from the US (most of my family has been in the US since colonial times). There is also a Canary Island ancestor from the 1700s on my mother's side. My mom came up as .1% N. African and .5% Sub-Saharan (the sub being too high to be explained by Canarian -I think natives show like 2% SSA?). My dad seems to have about .8% SSA. His dad's mom was adopted so I don't know anything about her.

What do you think about the E. European and the African given my gedmatch results? I really appreciate any insight you can provide.

Davidski said...

Well, almost all Europeans show some "East European" in the EUtest, and most score quite a bit of it. However, membership in this cluster is only relevant in terms of recent ancestry if it has a clear influence on your Oracle lists. In other words, if you have recent Eastern European ancestry then you'll see Eastern European countries and ethnic groups popping up regularly in your Oracle results. But even so, this could well be due to German ancestry, especially if your German ancestors came from east of the Elbe.

On the other hand, Europeans see very little or no Sub-Saharan admixture in the EUtest, so if you're getting around 1% then that's fairly unusual and it probably represent a segment or two of real African admixture.

Davidski said...

It's possible to reconcile most of the different test results, unless they're actually erroneous.

That's because the same ancestry can be reported in many different ways. The reason for this is that genetic variation is clinal, and in these tests we're sort of pretending that it's made up of discrete clusters.

Natalie Baker said...

Hi David,
Thanks for your response. I do have a question about my Oracle results in the EU admix test, particularly for the secondary populations. As I mentioned, I do not know the origin of my grandfather's mother. I do know that my known German ancestors (on both sides) were from the Baden (SW) area. Then of course, lots of northern French, some Dutch, and the Canarian ancestor. Thus, I don't understand the sources of the secondary populations. If you could provide more insight about this I would appreciate it very much. Below are the Oracle results:

# Primary Population (source) Secondary Population (source) Distance
1 92.8% English + 7.2% Tuscan @ 1.88
2 95% English + 5% South_Italian_&_Sicilian @ 1.96
3 95.4% English + 4.6% GR @ 2.02
4 97% English + 3% Druze @ 2.03
5 96.8% English + 3.2% Armenian @ 2.04
6 96.9% English + 3.1% Assyrian @ 2.06
7 95.2% English + 4.8% AJ @ 2.06
8 97.2% English + 2.8% Samaritan @ 2.09
9 96.5% English + 3.5% TR @ 2.09
10 97.1% English + 2.9% Mandean @ 2.11
11 97% English + 3% Kurdish @ 2.16
12 97.1% English + 2.9% IQ @ 2.17
13 91.7% English + 8.3% PT @ 2.17
14 93.1% English + 6.9% North_Italian @ 2.18
15 97% English + 3% IR @ 2.18
16 96.7% English + 3.3% Moroccan @ 2.22
17 97% English + 3% Algerian @ 2.22
18 97.4% English + 2.6% GE @ 2.22
19 97.2% English + 2.8% Sardinian @ 2.23
20 97.5% English + 2.5% Mozabite_Berber @ 2.3

Davidski said...

It seems that your genome is a good fit for a mix made up of around 97% English (from Kent) and 3% Mediterranean or Near Eastern. So these results seem generally correct based on your reported ancestry, but obviously can't be taken too literally.

From memory there are different Oracles available on that EUtest page at GEDmatch. You should try and find one that breaks down the Western/Northwestern portion of your genome, and then see what else shows. But keep in mind there is no Canarian reference sample in the Oracle lists, so the algorithm will look for something close and that might also affect the rest of the results.

Natalie Baker said...

Right. This makes sense. I checked multiple oracles in various tools and I either come up as Italian, near East, or Lithuania, Romania, or Hungary on the secondary populations. If my great grandmother were to have origins in areas like the latter more E. European areas, would you expect to see results similar to mine? Thanks again for all your help.

Davidski said...

If your great grandmother was from Eastern Europe you'd be seeing countries and groups like Russia, Ukraine, Erzya, etc. regularly as secondary populations. You should also be seeing Sweden as a primary population regularly, and even Poland at times.

Natalie Baker said...

So the Lithuanian, Romanian, or Hungarian would confirm near East origins then -or more of the same of what you were previously saying? Sorry, I am just trying to understand.

Davidski said...

I don't think so. These look like signals that show up regularly across Western and Central Europe, probably due to massive migrations from east to west during the Bronze and Iron Ages. I believe you'd be seeing very different results if your great grandmother was from Eastern Europe.

Natalie Baker said...

Got it! Thanks again for all your help.

CatchingAllTheSouls said...

Davidski you have answered all of my questions and I really do appreciate it! However, I am looking at my oracle-4 results and I cannot tell if this is telling me that I am very mixed European or that it simply has no clue what I am. I am interpreting this as me being a solid mix of Anglo-Saxon, Native Briton, and maybe some Vikings from that bit of Swedish? What do you think of all of this?


Using 2 populations approximation:
1 50% Cornish +50% FR @ 4.086
2 50% FR +50% Scottish @ 4.241
3 50% FR +50% Orcadian @ 4.489
4 50% FR +50% IE @ 4.761
5 50% English +50% FR @ 4.982
6 50% ES +50% Scottish @ 5.308
7 50% FR +50% NL @ 5.383
8 50% PT +50% Scottish @ 5.471
9 50% ES +50% Orcadian @ 5.502
10 50% FR +50% West_&_Central_German @ 5.778
3081 iterations.



Using 3 populations approximation:
1 50% English +25% AT +25% French_Basque @ 3.579
2 50% West_&_Central_German +25% French_Basque +25% West_&_Central_German @ 3.858
3 50% West_&_Central_German +25% English +25% French_Basque @ 3.875
4 50% West_&_Central_German +25% French_Basque +25% NL @ 3.895
5 50% West_&_Central_German +25% Cornish +25% French_Basque @ 3.919
6 50% FR +25% Cornish +25% Scottish @ 3.951
7 50% NL +25% French_Basque +25% West_&_Central_German @ 4.006
8 50% Cornish +25% Cornish +25% ES @ 4.050
9 50% Cornish +25% FR +25% FR @ 4.086
10 50% FR +25% Cornish +25% Cornish @ 4.086
71107 iterations.



Using 4 populations approximation:
1 AT + English + English + French_Basque @ 3.579
2 AT + Cornish + English + French_Basque @ 3.786
3 AT + English + French_Basque + NL @ 3.815
4 FR + French_Basque + South_&_Central_Swedish + West_&_Central_German @ 3.841
5 French_Basque + West_&_Central_German + West_&_Central_German + West_&_Central_German @ 3.858
6 English + French_Basque + West_&_Central_German + West_&_Central_German @ 3.875
7 AT + English + French_Basque + West_&_Central_German @ 3.891
8 French_Basque + NL + West_&_Central_German + West_&_Central_German @ 3.895
9 Cornish + French_Basque + West_&_Central_German + West_&_Central_German @ 3.919
10 Cornish + FR + FR + Scottish @ 3.951
11 AT + English + French_Basque + Orcadian @ 4.003
12 French_Basque + NL + NL + West_&_Central_German @ 4.006
13 FR + French_Basque + NL + South_&_Central_Swedish @ 4.018
14 AT + Cornish + French_Basque + NL @ 4.031
15 English + French_Basque + NL + West_&_Central_German @ 4.045
16 Cornish + Cornish + Cornish + ES @ 4.050
17 English + FR + French_Basque + South_&_Central_Swedish @ 4.057
18 AT + Cornish + French_Basque + West_&_Central_German @ 4.085
19 Cornish + Cornish + FR + FR @ 4.086
20 FR + French_Basque + West_&_Central_German + West_&_Central_German @ 4.091

CatchingAllTheSouls said...

And here is the standard oracle:

Mixed Mode Population Sharing:

# Primary Population (source) Secondary Population (source) Distance
1 71.8% West_&_Central_German + 28.2% French_Basque @ 3.28
2 59.9% FR + 40.1% Scottish @ 3.47
3 57.9% Cornish + 42.1% FR @ 3.52
4 76.6% Cornish + 23.4% ES @ 3.55
5 73.8% NL + 26.2% French_Basque @ 3.64
6 77.8% Cornish + 22.2% PT @ 3.75
7 57.8% FR + 42.2% Orcadian @ 3.79
8 82.6% Cornish + 17.4% North_Italian @ 3.88
9 59.3% FR + 40.7% IE @ 3.98
10 64.3% Orcadian + 35.7% ES @ 3.98
11 61.8% Scottish + 38.2% ES @ 4.06
12 62.4% Scottish + 37.6% PT @ 4.12
13 93.2% Cornish + 6.8% Sardinian @ 4.14
14 87.4% Cornish + 12.6% Tuscan @ 4.24
15 75.6% English + 24.4% French_Basque @ 4.27
16 85.4% Cornish + 14.6% French_Basque @ 4.31
17 65.4% Orcadian + 34.6% PT @ 4.34
18 54.7% FR + 45.3% English @ 4.35
19 68.7% English + 31.3% ES @ 4.4
20 63.7% IE + 36.3% ES @ 4.44

Thinking... said...

Hello and thank you very much for this indepth analysis. I won't pretend that I understand it all, i.e. K9 versus K11 etc. etc. in relation to my DNA. I ran the K9 so far.

I do have a question: I've noticed that the North Amerindian population results are from the Northwest and Arctic. Do you happen to know if any of your analyses include Northeastern Amerindian?

I had the ancestrybyDNA test done as well a while back and it showed 86% Caucasian and 14% Native American results. I have not been able to duplicate those results with AncestryDNA. My ancestor was from New York who falls within the Northeast Amerindian population.

Thanks in advance. I've entered my AncestryDNA results into the gedmatch database.

Davidski said...

There's really isn't much of a difference between Native American groups in this context. In other words, using reference samples from Northwestern or even Central America should show the right amount of native admixture from Northeastern America.

Keep in mind that the AncestrybyDNA test used a very small number of markers. I could be wrong, but I seem to recall it was around 100. Such small panels of markers can result in inflated admixtures for some individuals.

So if you can't reproduce the 14% figure, or thereabouts, with any sort of high density test, like those at Ancestry.com or 23andMe, then you simply don't have that much Amerindian ancestry.

Thinking... said...

Thank you very much for explaining this to me. Since it was my great grandmother who was an Onandago from New York, when I received my results I thought that the percentage was awfully high. My brother used the same testing company and his results were 19% Native American. We weren't asked any questions about what we were looking for to influence the results. It actually surprised us that we received those percentages. Incidentally, my brother also did the AncestryDNA along with me and his results were a bit different than mine; him receiving 99% British Isles and me receiving 84% British Isles and 14% Central Europe. We are definitely full siblings and it just goes to show the different in admixture inheritance within one family. Again, thank you for your help.

Gail said...

Hello Davidski,

Very new to this. My son is adopted from Guatemala. According to Dr. McDonald results, he is almost 95% Maya. Can you tell me which of your tests might be the most useful for us? It looks like K9, K10, K11,K12 all have North Amerindian + Arctic - Northwest America listed. Thanks for everythin!

Corey said...

Good Evening Davidski,
Thank you for your informative site. I am also new to this. I have read through most of your gedmatch linked blogs and have learned multitudes of information, however I have a few questions. Here is my info from EUTEST Oracle 4 Ancestors;
Kit Number: A014423
# Population Percent
1 ATLANTIC 24.03
2 NORTH-CENTRAL_EURO 23.93
3 WEST_MED 12.78
4 SOUTH_BALTIC 11.66
5 EAST_EURO 7.84
6 EAST_MED 7.74
7 WEST_ASIAN 7.28
8 MIDDLE_EASTERN 3.39
9 SOUTH_ASIAN 1.34

Using 1 population approximation:
1 West_&_Central_German @ 7.605
2 FR @ 8.122
3 NL @ 8.507
4 Cornish @ 9.314
5 English @ 9.478
6 AT @ 10.296
7 Orcadian @ 11.228
8 IE @ 11.671
9 DK @ 12.784
10 Scottish @ 12.966
78 iterations.

Using 2 populations approximation:
1 50% NO +50% North_Italian @ 4.857
2 50% North_Italian +50% South_&_Central_Swedish @ 5.087
3 50% DK +50% North_Italian @ 5.154
4 50% North_Italian +50% Orcadian @ 5.196
5 50% FR +50% West_&_Central_German @ 5.260
6 50% IE +50% North_Italian @ 5.277
7 50% North_Italian +50% Scottish @ 5.424
8 50% FR +50% NL @ 5.891
9 50% English +50% North_Italian @ 5.977
10 50% NO +50% Tuscan @ 6.119

Using 3 populations approximation:
1 50% IE +25% NL +25% South_Italian_&_Sicilian @ 2.805
2 50% IE +25% English +25% South_Italian_&_Sicilian @ 2.827
3 50% Orcadian +25% English +25% South_Italian_&_Sicilian @ 2.907
4 50% IE +25% South_&_Central_Swedish +25% South_Italian_&_Sicilian @ 2.944
5 50% Orcadian +25% NL +25% South_Italian_&_Sicilian @ 2.949
6 50% Orcadian +25% Orcadian +25% South_Italian_&_Sicilian @ 2.961
7 50% Orcadian +25% IE +25% South_Italian_&_Sicilian @ 2.980
8 50% IE +25% South_Italian_&_Sicilian +25% West_&_Central_German @ 3.001
9 50% IE +25% NO +25% South_Italian_&_Sicilian @ 3.025
10 50% IE +25% Orcadian +25% South_Italian_&_Sicilian @ 3.060
88634 iterations.

Using 4 populations approximation:
1 IE + IE + NL + South_Italian_&_Sicilian @ 2.805
2 English + IE + IE + South_Italian_&_Sicilian @ 2.827
3 English + IE + Orcadian + South_Italian_&_Sicilian @ 2.835
4 IE + NL + Orcadian + South_Italian_&_Sicilian @ 2.845
5 English + Orcadian + Orcadian + South_Italian_&_Sicilian @ 2.907
6 IE + IE + South_&_Central_Swedish + South_Italian_&_Sicilian @ 2.944
7 IE + NL + Scottish + South_Italian_&_Sicilian @ 2.945
8 NL + Orcadian + Orcadian + South_Italian_&_Sicilian @ 2.949
9 Cornish + IE + South_&_Central_Swedish + South_Italian_&_Sicilian @ 2.959
10 Orcadian + Orcadian + Orcadian + South_Italian_&_Sicilian @ 2.961
11 IE + Orcadian + Orcadian + South_Italian_&_Sicilian @ 2.980
12 NL + Orcadian + Scottish + South_Italian_&_Sicilian @ 2.996
13 IE + IE + South_Italian_&_Sicilian + West_&_Central_German @ 3.001
14 Cornish + IE + NO + South_Italian_&_Sicilian @ 3.020
15 IE + IE + NO + South_Italian_&_Sicilian @ 3.025
16 English + IE + Scottish + South_Italian_&_Sicilian @ 3.038
17 English + Orcadian + Scottish + South_Italian_&_Sicilian @ 3.057
18 IE + IE + Orcadian + South_Italian_&_Sicilian @ 3.060
19 IE + Orcadian + South_&_Central_Swedish + South_Italian_&_Sicilian @ 3.066
20 IE + Orcadian + South_Italian_&_Sicilian + West_&_Central_German @ 3.081

Oracle 4 seems to match my family oral history. I am of mixed European ancestry from America. Maternal side= Irish(Cork), Scottish (West Highlands), German. Paternal side= Sicilian(Palermo), Dutch, Irish. I love to study ancient cultures so this is very interesting for me.
Question 1. Is my admixture primarily Celt/Viking/Germanic/Med?
Question 2. Could Sicilian born grandfather have contributed French component? My last name, Chiaramonte is supposedly from Charlemagne's knight, Clermont...
Question 3. How much of my Sicilian dna contains Middle Eastern/North African? Palermo was a Muslim/Berber conquest at one point.
Thank you for your services.
Corey

Gail said...

My son's kit # is M174237

Davidski said...

Yes, any of the tests with Amerindian clusters should give him accurate ancestry proportions. There's also the Hunter-Gatherer vs. Farmer test which includes an Amerindian cluster based on South and Central American samples.

For a breakdown of European ancestry the EUtest is the best, although it uses Siberian and East Asian clusters as a stand-ins for Amerindian.

Davidski said...

Well I suppose Celt/Viking/Germanic/Med is one way to interpret your results. And yes, French might be showing up on your list because it's a convenient way for the algorithm to describe your mixed Northern European/Mediterranean ancestry. But it seems like an indirect result to me rather than due to an actual French ancestor.

I can't see any Middle Eastern or North African admixture in your output, but that doesn't mean your Sicilian ancestor didn't have such admixture. Exotic admixture is lost very quickly from generation to generation unless it's introduced into the pedigree regularly. Sicilians don’t usually carry more than a couple per cent of fairly recent Middle Eastern and North African admixture, so that can disappear from a pedigree very rapidly.

Corey said...

Thank you for the quick response. Ah, so the algorithm tries to connect the dots and for someone of my pedigree, France is the obvious default mixture. How would you interpret the results? My limited experience looking through all the gedmatch oracles etc...made me come to that simplified conclusion. The EUtest oracle 4 seemed to sum all the other tests in a general way. Time to study more!
Corey

William Boyce - Owner said...

My father is of British/German ancestry, and my mother is an Ashkenazi Jew. My father's Euro ancestry is discernable from the results, but the other 50% of my results (Ashkenazi) seem like a very broad mosaic. Any insights? And why am I getting Serbian, North Italian and Romanian 1-population matches?

Kit Number: M213921
Admix Results (sorted):

# Population Percent
1 ATLANTIC 17.85
2 NORTH-CENTRAL_EURO 16.40
3 EAST_MED 15.68
4 WEST_MED 14.49
5 EAST_EURO 10.29
6 SOUTH_BALTIC 9.92
7 MIDDLE_EASTERN 8.07
8 WEST_ASIAN 6.22
9 SOUTH_ASIAN 1.07


--------------------------------

Least-squares method.

Using 1 population approximation:
1 Serbian @ 10.654
2 North_Italian @ 11.345
3 RO @ 11.914
4 AT @ 13.651
5 PT @ 13.884
6 Tuscan @ 14.430
7 FR @ 15.384
8 HU @ 15.594
9 ES @ 16.266
10 West_&_Central_German @ 18.723
78 iterations.



Using 2 populations approximation:
1 50% AJ +50% NL @ 4.080
2 50% AJ +50% Cornish @ 4.209
3 50% AJ +50% West_&_Central_German @ 4.287
4 50% AJ +50% DK @ 4.834
5 50% AJ +50% Orcadian @ 4.978
6 50% AJ +50% English @ 4.999
7 50% North_Swedish +50% South_Italian_&_Sicilian @ 5.087
8 50% AJ +50% IE @ 5.261
9 50% AJ +50% AT @ 5.432
10 50% AJ +50% NO @ 5.450
3081 iterations.



Using 3 populations approximation:
1 50% AJ +25% FR +25% North_Swedish @ 2.815
2 50% FR +25% Samaritan +25% South_Finnish @ 2.833
3 50% ES +25% Mandean +25% South_Finnish @ 2.891
4 50% AJ +25% ES +25% North_Swedish @ 2.924
5 50% AJ +25% ES +25% South_Finnish @ 3.008
6 50% FR +25% East_Russian +25% Samaritan @ 3.083
7 50% ES +25% IQ +25% South_Finnish @ 3.165
8 50% FR +25% East_Finnish +25% Samaritan @ 3.177
9 50% FR +25% Samaritan +25% West_Russian @ 3.286
10 50% PT +25% Mandean +25% South_Finnish @ 3.308
95631 iterations.



Using 4 populations approximation:
1 AT + ES + North_Swedish + Samaritan @ 2.538
2 ES + HU + NO + Samaritan @ 2.764
3 AJ + AJ + FR + North_Swedish @ 2.815
4 FR + FR + Samaritan + South_Finnish @ 2.833
5 DK + FR + HU + Samaritan @ 2.858
6 FR + HU + IE + Samaritan @ 2.878
7 AT + ES + NO + Samaritan @ 2.887
8 ES + ES + Mandean + South_Finnish @ 2.891
9 ES + IE + Samaritan + Ukrainian-Russian @ 2.907
10 DK + ES + HU + Samaritan @ 2.918
11 AJ + AJ + ES + North_Swedish @ 2.924
12 ES + Mandean + PT + South_Finnish @ 2.956
13 Cornish + East_Russian + ES + Samaritan @ 2.970
14 AT + ES + Samaritan + South_&_Central_Swedish @ 2.978
15 English + ES + Samaritan + West_Russian @ 2.983
16 Cornish + ES + Samaritan + West_Russian @ 2.984
17 Cornish + ES + Samaritan + Ukrainian-Russian @ 2.989
18 AJ + AJ + ES + South_Finnish @ 3.008
19 ES + FR + Samaritan + South_Finnish @ 3.016
20 Cornish + FR + HU + Samaritan @ 3.020

275023 iterations.

Done.

Elapsed time 0.1827 seconds.

Davidski said...

It's probably because 50% British/German and 50% AJ looks more or less like Serbian, North Italian or Romanian to the algorithm.

William Boyce - Owner said...

By eliminating the results that are likely proxies for my fathers British/German ancestry, and then looking at only the likely AJ proxies, I get:

HU + Samaritan @ 2.764
HU + Samaritan @ 2.858
HU + Samaritan @ 2.878
Samaritan + Ukrainian-Russian @ 2.907
HU + Samaritan @ 2.918
East_Russian + Samaritan @ 2.970
Samaritan + West_Russian @ 2.983
Samaritan + West_Russian @ 2.984
Samaritan + Ukrainian-Russian @ 2.989
HU + Samaritan @ 3.020

The Samaritan results are clear and fit PERFECTLY with Ashkenazi Eastern Med origins.

The other results are a split between HU and Russian. Would this be in line with Eastern European admixture in my Ashkenazi line?



Davidski said...

Yeah, I think you hit the nail on the head there. Samaritan plus Hungarian or East Slavic sounds like Ashkenazi to me, more or less.

Corey said...

Hello David,
Corey here, A014423. How would you put my admixture in percentages using the oracles? Ancestry has me at 70% British Isles, 15% Persian/Turk/Cauc and 9% East. european with 6% unknown. I have played around with the different oracles and have become more confused because different oracles are using different population bases perhaps. This is why I used simpler terms like Celt/viking/Germanic/Med. in an earlier post Thanks for the help.
Corey

Davidski said...

I think that overall your description was accurate. The point I was making was that there are many different ways to say the same thing.

who are you said...

David, do you think it is plausible to create a GEDmatch tool that can provide Oracle-like results for chromosome segments? I think this could be useful if you find someone you share a segment with because that would provide defining boundaries for the segment. I do some ancestry painting gazing for segments, but I just don't have enough resources to know what reference populations could be represented when there are multiple colors/populations.

I guess it could still be misrepresentative since the segment would only represent a small percentage of the original donor, but it just would be cool to be able to pin the segments to specific people in the pedigree.

Edek said...

Hi, David!

Edek here, F291701. First of all, I would like to congratulate and thank you for your hard and excellent work, helping people with Gedmatch. It's really priceless.

I just got my Family Finder results from FTDNA and I'm a little bit confused. Well, that's what I knew from my patrilineal and matrilineal lines before taking the test, based on my 8 great grandparents.

*Father's side: 3/4 Portuguese (most of them from Azores); 1/4 (black/mulatto).

*Mother's side: 1/2 (mixed ancestry Portuguese, German and Native American); 1/4 Polish/Czech (Central Poland, Galicia and Brno region); 1/4 Italian (Veneto).

These are my FF results:
*Europe 41,84% (Finnish, French, Orcadian and Spanish);
*Native American 6,56% (Surui, Pima, Columbian, Maya);
*Middle East 37,87% (exclusively Jewish);
*West Africa 13,74% (mandenka, yoruba).

It's hard for me to fit some pieces of this puzzle. I don't understand why Orcadians, French and Finnish showed up in my results as well as from where I would have those 37,87% as middle eastern Jew.

What do you think about these results? Which of the Gedmatch tests would give me a more accurate results?

Davidski said...

Test outcomes are always limited by the reference used. If, say, Orcadians are the only reference set used to represent 1/4 of Europe (in this instance, Northwest, North and even Central Europe) then you’re bound to see them in your results if you're even partly Northern or Central European.

Try the EUtest and Jtest, and all the Oracles linked to those tests. Just keep in mind that these analyses are designed for native Europeans with no recent admixture from outside of Europe. So in your case Amerindian admixture will be represented by Siberian admixture, and you're likely to see northeast Siberian reference samples like the Chukchi in your Oracle results.

But the most accurate tests for you in regards to ancestral proportions will be the Eurogenes K9 and the Hunter-Gatherer vs. Farmer test.

Edek said...

Thank you, once again!

Ruben astro said...

Hey Davidski,

These results were really interesting.
What would you guess where I was from?
I'm a mix between two different races but both European.

Population
SOUTH_BALTIC 6.85%
EAST_EURO 4.91%
NORTH-CENTRAL_EURO 20.03%
ATLANTIC 23.19%
WEST_MED 14.30%
EAST_MED 18.60%
WEST_ASIAN 9.46%
MIDDLE_EASTERN 2.66%
SOUTH_ASIAN -
EAST_AFRICAN -
EAST_ASIAN -
SIBERIAN -
WEST_AFRICAN -


Admix Results (sorted):

# Population Percent
1 ATLANTIC 23.19
2 NORTH-CENTRAL_EURO 20.03
3 EAST_MED 18.60
4 WEST_MED 14.30
5 WEST_ASIAN 9.46
6 SOUTH_BALTIC 6.85
7 EAST_EURO 4.91
8 MIDDLE_EASTERN 2.66


--------------------------------

Least-squares method.

Using 1 population approximation:
1 North_Italian @ 10.247
2 Tuscan @ 11.519
3 PT @ 13.627
4 FR @ 13.873
5 ES @ 15.133
6 Serbian @ 16.937
7 AT @ 17.357
8 RO @ 17.377
9 West_&_Central_German @ 18.880
10 NL @ 20.122
78 iterations.



Using 2 populations approximation:
1 50% Scottish +50% South_Italian_&_Sicilian @ 5.018
2 50% Cornish +50% South_Italian_&_Sicilian @ 5.297
3 50% IE +50% South_Italian_&_Sicilian @ 5.311
4 50% GR +50% Scottish @ 5.315
5 50% Orcadian +50% South_Italian_&_Sicilian @ 5.455
6 50% English +50% South_Italian_&_Sicilian @ 5.524
7 50% GR +50% IE @ 5.707
8 50% Cornish +50% GR @ 5.764
9 50% GR +50% Orcadian @ 5.965
10 50% English +50% GR @ 6.040
3081 iterations.



Using 3 populations approximation:
1 50% FR +25% Druze +25% Scottish @ 4.920
2 50% Scottish +25% GR +25% South_Italian_&_Sicilian @ 4.956
3 50% South_Italian_&_Sicilian +25% English +25% Scottish @ 4.970
4 50% FR +25% Druze +25% IE @ 4.972
5 50% South_Italian_&_Sicilian +25% Cornish +25% Scottish @ 4.989
6 50% Scottish +25% South_Italian_&_Sicilian +25% South_Italian_&_Sicilian @ 5.018
7 50% South_Italian_&_Sicilian +25% Scottish +25% Scottish @ 5.018
8 50% South_Italian_&_Sicilian +25% IE +25% Scottish @ 5.128
9 50% West_&_Central_German +25% Druze +25% French_Basque @ 5.131
10 50% English +25% Druze +25% French_Basque @ 5.148
106287 iterations.

Parkside 1503 said...

I am an admix W. African/European. I also was told I had Native American in me both maternally and paternally.

Having done Ancestry's mtDNA in 2008, I was identified in Haplogroup X. Then as things progressed, I was indicated as an X2. When I inquired later, Ancestry re-tested me and identified me as a L1c2a1. For some reason, I found the fact that they retested me and identified me as a new Haplogroup was troubling. I could reason that as tests improve with accuracy, there would be some reasonable shift, a greater focus on more specific DNA information (i.e. Region,as indicated by your tests), but there was something amiss with this new identity. It was not wholly congruent with family history, both orally, or with research findings. Now, I took the Autosomal DNA, and discovered more accuracy with my family history. 56% W.African, 22% Scandinavian, 18% British Isles and 4% Unknown.

Fast-forward to my results in K-9/K12/K13, it confirms without a doubt that my DNA makeup parallels exactly with the Haplogroup X2 migratory path. Funny how 4% unknown can really get a person riled up. LOLOL!!! Once again, this is such a valuable test and I truly thank you. Being multi-racial helps me in identifying factors for research with my White relatives as well as for my African American relatives. Thanks once again.

Jared Knows said...

Hello David, I am new to all of this and I was just wondering if I could send you some of my results from your programs and see what you think.I know you are probably very busy and there is a lot I'm unsure about. Thanks

Davidski said...

Unfortunately I can't accept any raw data at the moment. But if you start slowly with the tools at GEDmatch, using the posts and comments here as a guide, you'll be an expert in interpreting your data in no time.

Unknown said...

Not sure what to make of the results i got from Eurogenes and Dodecad. FTDNA says I'm Middle East (Jewish), and JEWISH in big bold letters at over 35% with like a 0.3 margin of error. Well according to JTest i'm only 2.21 Ashkenaz. I realize other factors come in to play, like if you add my South Baltic, with both Mediterraneans, and the Ashkenazi, you get around 35%. So it would seem that my influence might be Sephardic? especially with the mediterranean, and that fact that my ancestors on my fathers side were Portuguese. HOWEVER, my mothers maternal side at 2nd Great Grandparents were from Germany/Prussia......all of them.....and they all have Jewish names. So i'm a little bit baffled at the results i'm getting.


Population
SOUTH_BALTIC 8.61%
EAST_EURO 12.62%
NORTH-CENTRAL_EURO 20.98%
ATLANTIC 20.97%
WEST_MED 12.40%
ASHKENAZI 2.21%
EAST_MED 6.93%
WEST_ASIAN 1.23%
MIDDLE_EASTERN 0.83%
SOUTH_ASIAN 1.58%
EAST_AFRICAN -
EAST_ASIAN 3.79%
SIBERIAN 6.18%
WEST_AFRICAN 1.68%

Tim Dennis said...

Hi, I have African decent. Look at this results. I'm trying to find a Native American link, but I don't see anything.

SOUTH_BALTIC 2.51%
EAST_EURO 4.25%
NORTH-CENTRAL_EURO 9.32%
ATLANTIC 4.58%
WEST_MED 5.07%
ASHKENAZI 2.19%
EAST_MED 2.53%
WEST_ASIAN 2.58%
MIDDLE_EASTERN -
SOUTH_ASIAN 0.33%
EAST_AFRICAN 14.31%
EAST_ASIAN -
SIBERIAN 0.73%
WEST_AFRICAN 51.60%

Davidski said...

You're using the wrong test. The Jtest doesn't have an Amerindian cluster.

For the most accurate Native American results try the Eurogenes K9 or K10, and then make sure to have a look at the chromosome painting, which will show you if you have any large Amerindian segments.

Davidski said...

The South Baltic component isn't a very useful measure of Jewish ancestry, because most Jews are essentially an East Mediterranean people. Only Jews with recent ancestry from Northern Europe will show a lot of South Baltic.

If you really do have recent Ashkenazi ancestry, then you should be seeing AJ in the Jtest Oracle results. If you're not, then yes, perhaps your ancestors were Sephardic Jews. But if so, then you should be seeing significant influence from the East Mediterranean and the Near East in your Admix and Oracle results.

The figures you posted above don't indicate any Ashkenazi ancestry, and perhaps minimal Sephardic ancestry, depending on what your other ethnic origins are. So I have no idea why FTDNA has labeled you Jewish with big bold letters? You need to investigate that further.

g and r said...

I curious as to why my Ashkenazi on Jtest to Oracle X jumped from 3.42 to 8.16? Is this enough to substantiate a closer generation? The Mixes seem about right with what we know, mothers both sides Swedish to at least 1600, fathers side eastern European/Russian/northern Poland (always the rumour of Jewish ancestor. Also a chance of Scandinavian on fathers side due to marauding Vikings I guess!

Kit Number: M045540

Admix Results: Jtest and Oracle X
# Population Percent
1 SOUTH_BALTIC 17.85
2 EAST_EURO 19.92
3 NORTH-CENTRAL_EURO 23.71
4 ATLANTIC 21.80
5 WEST_MED 7.70
6 ASHKENAZI 3.42
7 EAST_MED 3.22
8 WEST_ASIAN 2.36
9 MIDDLE_EASTERN 0.00
10 SOUTH_ASIAN 0.00
11 EAST_AFRICAN 0.00
12 EAST_ASIAN 0.00
13 SIBERIAN 0.00
14 WEST_AFRICAN 0.00


Pct. Calc. Option 2

0 Unable to determine 0.02%
1 North_Swedish 45.73%
2 AJ 8.16%
3 West_Russian 6.97%
4 RO 6.89%
5 IE 6.51%
6 Serbian 6.01%
7 Ukrainian-Russian 5.60%
8 South_Finnish 5.46%
9 NO 5.28%
10 French_Basque

EU Test and Oracle X

Admix Results:

# Population Percent
1 SOUTH_BALTIC 17.88
2 EAST_EURO 20.20
3 NORTH-CENTRAL_EURO 23.92
4 ATLANTIC 22.23
5 WEST_MED 8.13
6 EAST_MED 4.37
7 WEST_ASIAN 2.91
8 MIDDLE_EASTERN 0.36
9 SOUTH_ASIAN 0.00
10 EAST_AFRICAN 0.00
11 EAST_ASIAN 0.00
12 SIBERIAN 0.00
13 WEST_AFRICAN 0.00


Pct. Calc. Option 2

1 North_Swedish 62.72%
2 RO 9.91%
3 AT 8.61%
4 West_Russian 7.76%
5 HU 5.44%
6 French_Basque 3.02%
7 Cornish 1.13%
8 North_Russian 1.09%
9 Serbian 0.31%
10 Samaritan

Here's my 23andMe breakdown

From father

8.8%
French and German
4.2%
Scandinavian
1.0%
Finnish
0.7%
British and Irish
12.4%
Nonspecific Northern European
13.1%
Eastern European
4.3%
Ashkenazi

Southern European
0.2%
Iberian
0.2%
Nonspecific Southern European
5.1%
Nonspecific European

From mother

24.0%
Scandinavian
5.4%
Finnish
2.7%
British and Irish
0.6%
French and German
16.3%
Nonspecific Northern European
0.3%
Eastern European
0.1%
Ashkenazi
0.6%
Nonspecific European
< 0.1%
Sub-Saharan African
< 0.1%
Unassigned

David Joyce said...
This comment has been removed by the author.
David Joyce said...

Hello again,

Sorry for the many questions thus far. I just have one more :} What populations are included in the Atlantic section for the eutest?

Thanks much!

Davidski said...

The Atlantic component is found all over Europe, and even in North Africa and West Asia. But it peaks on the Atlantic coast of Europe, especially in western Ireland and in Basque country. You can see that on this map, which shows the Jtest components, but these are basically the same as the EUtest components.

http://imageshack.us/a/img594/6464/jtestmaps.png

David Joyce said...

ah interesting! Would this explain my SNP group:
R-SRY2627?

thanks again!

Davidski said...

Yes, I'd say your high autosomal Atlantic affinity and Y-chromosome haplogroup correlate very well.

David Joyce said...

Hello,

What do the PCA numbers stand for?

Sincerely,

David

Courtney Creates said...

Hello David! I am a newbie to this DNA stuff and at this point I am pretty confused. I know from paper trail that most of my ancestors came from England/Scotland/Wales, Ireland, Germany/Prussia, and my paternal grandma always said her grandma was 100% Ojibway. I started DNA testing to try to prove this. I took AncestryDNA and uploaded my results to ftdna (waiting on results) and gedmatch. I ran the EUtest you mention to others above and the oracle. Here is what it said.
EUtest Oracle
Kit A101914
K=20
MinCount=2
Admix results sorted
# Population Percent
1 NORTH-CENTRAL_EURO 29.01
2 ATLANTIC 27.34
3 WEST_MED 14.33
4 EAST_EURO 12.53
5 SOUTH_BALTIC 10.58
6 WEST_ASIAN 2.67
7 MIDDLE_EASTERN 1.89
8 SOUTH_ASIAN 1.02
9 EAST_MED 0.63

Single Population Sharing:

# Population (source) Distance
1 Cornish 4.95
2 English 5.12
3 NL 5.84
4 IE 5.87
5 Orcadian 5.92
6 DK 6.19
7 West_&_Central_German 7.51
8 Scottish 7.52
9 NO 8.12
10 South_&_Central_Swedish 9.4
11 FR 10.83
12 North_Swedish 12.96
13 AT 14.27
14 PT 17.71
15 ES 17.83
16 HU 19.41
17 South_Finnish 21.55
18 Serbian 21.63
19 French_Basque 22.48
20 North_Italian 22.64

Mixed Mode Population Sharing:

# Primary Population (source) Secondary Population (source) Distance
1 81.4% DK + 18.6% French_Basque @ 3.52
2 75.6% NO + 24.4% French_Basque @ 3.86
3 91.2% DK + 8.8% Sardinian @ 4.1
4 79.6% DK + 20.4% ES @ 4.32
5 92.5% Orcadian + 7.5% Sardinian @ 4.44
6 66.5% Cornish + 33.5% DK @ 4.46
7 69.8% DK + 30.2% FR @ 4.48
8 72.7% South_&_Central_Swedish + 27.3% French_Basque @ 4.49
9 92.8% IE + 7.2% Sardinian @ 4.54
10 79% Cornish + 21% NO @ 4.62
11 91% English + 9% French_Basque @ 4.64
12 72.1% NO + 27.9% ES @ 4.66
13 80.9% DK + 19.1% PT @ 4.68
14 95.9% English + 4.1% Sardinian @ 4.68
15 56.5% Cornish + 43.5% English @ 4.7
16 96.4% English + 3.6% Mozabite_Berber @ 4.74
17 86.2% Cornish + 13.8% South_&_Central_Swedish @ 4.78
18 97.7% Cornish + 2.3% Mozabite_Berber @ 4.81
19 65.9% English + 34.1% IE @ 4.81
20 96.2% English + 3.8% Moroccan @ 4.84

What does this mean? Other k9 results said I had some america-indian does this mean she was right? I am confused.

Davidski said...

You'd probably be showing some Siberian admix in the EUtest if you had Native American ancestry. The fact that Mozabite Berber and Moroccan showed up in the oracle results might suggest some very minor African-American ancestry. But it's hard to tell from that. You'd have to study your chromosome paintings and cross check the results.

Courtney Creates said...

What does NL, DK, and NO stand for on the oracle admixture test?

Ruben astro said...

NL =The Netherlands, Dk=Denmark? NO=Norwich ?

Davidski, the results of the new EUtest are a bit different.
Now the 50% southern_Italian is transformed into Tuscan!
All other calculators are placing me as at least 50 % Southern Italian rather than Tuscan at first or second.

What reference population do you use under the name 'Tuscan' ?

keefd333 said...

Hi David,

I am really hoping you can help. I have ran just about every test and am not seeing any Native American. However I am seeing South and West Asian. Could this be the Native American? My grandfather was 1/4 Cherokee Indian (there is no doubt as his mother lived on a reservation). The rest of my ancestry is a mix of European as my ancestors have pretty much been in N. America since the late 1600s. I would love to get your insight and help putting my ancestry in perspective.

My kit number is M210016

Thanks,
Keith Andrews

Davidski said...

The South and West Asian components are from your European ancestors.

If you're 1/16 Native American, that should show up in this test as minor Amerindian in the admix results, and minor North Amerindian in the oracle results.

Wyatt said...

Hi,

I hope you can help me sort out my results. I am showing Dutch high on my results when the only Dutch ancestor I have is from around 1750. I also have a good bit of German ancestors from about 1650. My most recent ancestry is from Scotland, England, Ireland, and Poland. With that in mind, I would think I would have the British Isles and Western Europe on top, but I don't. If you could look at my data and give me your opinion, I would appreciate it very much.
Thanks,
Wyatt Lane


Eurogenes EUtest V2 K15 Oracle results:
Kit A683910

Admix Results (sorted):


# Population Percent
1 North_Sea 37.44
2 Atlantic 25.5
3 Baltic 11.77
4 Eastern_Euro 11.03
5 West_Med 7.99
6 East_Med 2.47
7 South_Asian 1.18
8 Oceanian 0.63
9 Siberian 0.6
10 Amerindian 0.56
11 Northeast_African 0.47
12 Southeast_Asian 0.18
13 West_Asian 0.16

Single Population Sharing:


# Population (source) Distance
1 Dutch 3.3
2 Norwegian 4.07
3 Orcadian 4.31
4 Danish 4.56
5 Swedish 5.69
6 West_Scottish 5.76
7 Southeast_English 5.78
8 Irish 6.14
9 Southwest_English 6.38
10 German 6.85
11 North_Swedish 7.81
12 French 13.02
13 Southwest_Finnish 13.55
14 Finnish 16.52
15 Hungarian 16.61
16 Austrian 17.08
17 Spanish_Galicia 20.36
18 Spanish_Cataluna 20.61
19 Serbian 21.14
20 East_Finnish 21.32

Mixed Mode Population Sharing:


# Primary Population (source) Secondary Population (source) Distance
1 88.2% Norwegian + 11.8% Spanish_Aragon @ 2.25
2 88% Norwegian + 12% Spanish_Valencia @ 2.31
3 90.1% Norwegian + 9.9% French_Basque @ 2.32
4 87.4% Norwegian + 12.6% Spanish_Murcia @ 2.36
5 88.3% Norwegian + 11.7% Spanish_Castilla_La_Mancha @ 2.39
6 87.6% Norwegian + 12.4% Spanish_Cantabria @ 2.41
7 87.8% Norwegian + 12.2% Southwest_French @ 2.41
8 86.3% Norwegian + 13.7% Spanish_Cataluna @ 2.44
9 89.1% Norwegian + 10.9% Spanish_Andalucia @ 2.47
10 87.2% Norwegian + 12.8% Spanish_Castilla_Y_Leon @ 2.5
11 87.8% Norwegian + 12.2% Spanish_Extremadura @ 2.53
12 87.2% Norwegian + 12.8% Portuguese @ 2.57
13 62.7% Norwegian + 37.3% Southeast_English @ 2.7
14 87% Norwegian + 13% Spanish_Galicia @ 2.74
15 65.7% Norwegian + 34.3% Southwest_English @ 2.74
16 87.2% Orcadian + 12.8% Ukrainian @ 2.83
17 81.3% Norwegian + 18.7% French @ 2.83
18 55.1% Norwegian + 44.9% Danish @ 2.88
19 56% Southwest_English + 44% North_Swedish @ 2.89
20 89.4% Orcadian + 10.6% Kargopol_Russian @ 2.89

Davidski said...

Well, if your recent ancestry is from Scotland, England, Ireland, and Poland, then you're likely to get Dutch as your top match because you're being pulled just to the east of the UK by your Polish ancestry. Of course, the first country just east of the UK is The Netherlands. After that it's Northwestern Germany and Scandinavia. That's why you're showing so much Norwegian in the results.

Sam Hilsen said...

Hi David,

Just got my results and am slightly confused to how to interpret them. The tests all show around ~0.5% Amerindian ancestry and ~1.5% African ancestry, is that just noise, or is that something tangible? Does the constantly high Caucasian admixture (over 25%) suggest a Caucasian ancestor? If it's not too much trouble I'd appreciate your insight :)

My kit number is M392217

Davidski said...

0.5% Amerindian is well within noise range, but 1.5% Sub-Saharan is well above the noise range and most likely a sign of fairly recent African ancestry. 25% Caucasian is way too high for Northern and Eastern Europeans, but suggests either recent Southern European and/or Jewish ancestry. You should try the new Eurogenes K13 oracles to see what sort of ethnic groups appear in your results.

By the way, I can't access your results with that kit number, because I can't log in as you.

Sam Hilsen said...

Thank you. Is there a way you could log in as me, if I give you my account details? I don't consider myself to be a Northern or Central European, ethnically. Some Northern and Central ancestry, sure, probably. I have Jewish heritage, but wouldn't that turn up as Southwest Asian or Middle Eastern, where Middle Eastern is an option? K13 comes up as East Med, West Med, North Atlantic and West Asian, primarily, although everything but South Asian and Oceanian is represented.

Sam Hilsen said...
This comment has been removed by the author.
Davidski said...

I don't think it's a good idea for me to log into your GEDmatch account.

Anyway, your Jewish ancestry explains the high Caucasian scores and it probably explains your Sub-Saharan admixture, unless you have Colonial American ancestry, in which case that would be another likely option.

Don't forget to run all the different oracles linked to the K13. They should give you a lot of info about potential ethnic links you might have.

Sam Hilsen said...

Thank you :) Just by considering my K13 test, which was
34.19% MEDITERRANEAN
28.35% CAUCAUSUS
17.71% NORTHERN EUROPEAN
16.38% SOUTHWEST ASIAN
1.22% WEST AFRICAN
1.17% EAST ASIAN
0.64% NORTH AMERINDIAN & ARCTIC
0.20% SIBERIAN
0.13% SOUTH ASIAN
and thereby fairly typical of the other tests, is there anything you could tell me about my heritage? (The paper trail on my father's side goes back about 2.5 generations on average, so I'm really not 100% on things on that side of the family). Where would you 'place' those results (in terms of general ethnicity) or are they quite mixed?

Davidski said...

OK, so now try the different ethnic oracles linked to the test.

Sam Hilsen said...

Depending on the oracles, they place me either as Southern Italian/Sicilian, Ashkenazi Jewish, or Sephardi Jewish. There's a wide array of suggested mixed groups (apparently I'm 1.47 plates away from Lebanese Christian and Spaniard) on K13, which you said is the most accurate, other suggested mixings are 54-46 Tuscan/Ashkenazi - other mixes are Sephardic/Bulgarian Romanian, Austrian, Polish, Ukrainian and other East/Central European groups, Tuscan/Druze, and Druze/North Italian. (K13 suggests I'm closest to South Italians as a single group with a distance of 4.5. Is 4.5 fairly close, or fairly distant, and indicative of a mixed background?)

Davidski said...

Well, it seems you're mostly Jewish and Southern European, but also with some recent ancestry from other parts of Europe. It's difficult to get more specific than that because many ethnic groups are too similar genetically to be distinct from each other in these sorts of tests.

Sandra bertwell said...

I am so confused. I have ran every test. The Dodecad shows lots of Asian and Italian, others show french, others show native american and others show Afircan. It seems that I am either a great mix of many things or I am really confused as to what tests do what.

Tyler Jacobs said...

Hello,

My question is the MDLP and Dodecad admixture tests will not load for me.

I just uploaded my raw dna data a day or two ago so I didnt know if it took a while to recognize it not really sure about that?

Also my Mediterranean results were at about approximately 30% and was wondering what all countries that pertains too, I didn't know if Sardini was just a reference point for the center of the area or if it was ONLY Italy or not?

Like for instance, could it be surrounding countries of Italy? or by mediterranean does it mean anything in the mediterranean ocean?

Any feedback is much appreciated thank you! :)

Shirley said...

Hi, I've been using Eurocad. I would like to print out the picture. Is there anyway that I can get the entire image? It only seems to print the beginning section of the chromosome. Thanks. Shirley

Tory said...

Hi, Thank you for all of this great information. My family history is Irish on my Dad's side and German/Scottish/English on my Mom's father's side with Cherokee a few generations ago and mayan perhaps as well. the 23andme data run thru familytree dna yields 94% orcadian and 5.73 middle eastern.

I have run a few of the tests on genmatch and want to confirm that the NA does show in my DNA (1.7%). Also when I run the Jtest it shows 3.87% azkenazi. Can you interpret these results and/or suggest another test I should run. thank you!

Kit FB7666

Admix Results (sorted):
# Population Percent
1 North_Atlantic 47.54
2 Baltic 25.3
3 West_Med 12.81
4 West_Asian 5.54
5 East_Med 3.43
6 South_Asian 1.66
7 Siberian 1.27
8 Amerindian 1.16
9 Red_Sea 0.93
10 Northeast_African 0.36

Single Population Sharing:
# Population (source) Distance
1 Dutch 2.07
2 German 3.56
3 Orcadian 3.69
4 Danish 3.79
5 Southeast_English 4.18
6 Irish 4.85
7 Norwegian 5.47
8 West_Scottish 5.53
9 Southwest_English 5.65
10 Swedish 6.58
11 North_Swedish 11.83
12 French 11.97
13 Austrian 14.53
14 Hungarian 15.55
15 Southwest_Finnish 19.39
16 Spanish_Cataluna 19.51
17 Southwest_French 20.55
18 Spanish_Galicia 20.55
19 Spanish_Castilla_Y_Leon 20.62
20 Portuguese 20.94

Mixed Mode Population Sharing:
# Primary Population (source) Secondary Population (source) Distance
1 97.9% Dutch + 2.1% Chukchi @ 1.13
2 98% Dutch + 2% Koryak @ 1.16
3 98% Dutch + 2% East_Greenlander @ 1.22
4 81.7% Orcadian + 18.3% Hungarian @ 1.26
5 69.1% Norwegian + 30.9% French @ 1.26
6 98.3% Dutch + 1.7% Evens @ 1.28
7 97.8% Dutch + 2.2% West_Greenlander @ 1.3
8 74.2% West_Scottish + 25.8% Hungarian @ 1.31
9 98.2% Dutch + 1.8% Dolgan @ 1.31
10 98.2% Dutch + 1.8% Yakut @ 1.31
11 98.3% Dutch + 1.7% Evenki @ 1.31
12 98.3% Dutch + 1.7% Oroqen @ 1.36
13 80.9% Orcadian + 19.1% Austrian @ 1.37
14 97.9% Dutch + 2.1% Ket @ 1.37
15 98% Dutch + 2% Tuvinian @ 1.38
16 98.1% Dutch + 1.9% Buryat @ 1.38
17 97.5% Dutch + 2.5% Shors @ 1.39
18 97.8% Dutch + 2.2% Altaian @ 1.39
19 81.1% Norwegian + 18.9% Spanish_Extremadura @ 1.4
20 98% Dutch + 2% Selkup @ 1.43


Jtest
Population
SOUTH_BALTIC 12.78%
EAST_EURO 14.73%
NORTH-CENTRAL_EURO 29.00%
ATLANTIC 23.41%
WEST_MED 11.26%
ASHKENAZI 3.87%
EAST_MED 0.97%
WEST_ASIAN 3.08%
MIDDLE_EASTERN 0.10%
SOUTH_ASIAN 0.79%
EAST_AFRICAN -
EAST_ASIAN -
SIBERIAN -
WEST_AFRICAN -

HarappaWorld Oracle results:
23 April 2013 - Oracle reference population percentages revised.

Kit FB7666
Admix Results (sorted):
# Population Percent
1 NE-Euro 50.97
2 Mediterranean 31.07
3 Baloch 9.95
4 Caucasian 5.85
5 American 1.11
6 SW-Asian 0.56
7 Beringian 0.5

Lasse Yisrael Rosenberg said...

Hi, Davinski
I,am of the Norwegian Romanipepole and little Jewish ancestery , can you pleas tell me watt Cain of test
is good fore me.

Rosenberg

Davidski said...

The K13 plus the oracles.

JSclark said...

Hello,
I am trying to determine more about my Native American ancestry and possible continents of origin. My kit is M144348. I ran some of the tests but without knowing which one to look at specifically, they all just look different to me. What test is most specific and what is it telling me? Thank you.

Davidski said...

The K13 will be most useful in your case, but just make sure you study the chromosome paintings and oracle results in detail. Look for Amerindian as well as Siberian chromosome segments, because North Amerindians come out significantly Siberian in this test.

Tory said...

Hi Davidski, I am also trying to resolve family history about my great great grandmother - oral history indicates she was native american, and some of my individual chromes record show a fairly high percentage of NA but the admixtures alone are somewhat low - I also have some reason to believe there is african american blood as well on my mothers' side. Would you comment on these results in that context? Thanks!

Admix Results (sorted):
# Population Percent
1 North_Atlantic 47.54
2 Baltic 25.3
3 West_Med 12.81
4 West_Asian 5.54
5 East_Med 3.43
6 South_Asian 1.66
7 Siberian 1.27
8 Amerindian 1.16
9 Red_Sea 0.93
10 Northeast_African 0.36

Population
North_Sea 37.09%
Atlantic 24.39%
Baltic 10.32%
Eastern_Euro 11.06%
West_Med 9.35%
West_Asian 3.32%
East_Med 1.09%
Red_Sea 0.29%
South_Asian 1.38%
Southeast_Asian -
Siberian 0.45%
Amerindian 0.89%
Oceanian -
Northeast_African 0.35%
Sub-Saharan -

Davidski said...

There seems to be minor Amerindian ancestry there, expressed as both Amerindian and Siberian admixture. But the African admixture looks like noise, because it's very low and showing up as Northeast African, which is not something you're likely to get from an African-American ancestor, but rather a Jewish or Southern European one. Although like I say, the level is too low to consider it as anything but noise.

Tory said...

Thank you for your evaluation Davidski!

Shuffle said...

Good evening,

I come from making the EUtest V2 K15-Test and I think my results turn me out as a very typical South-German, all my ancestors came from regions as Bavaria, Bade-Wurttemberg, Alsace, Lorraine, Rhénanie-Palatine etc. what surprises me, is the Oceanian component, can this be a calculation error? It is not very high and my family lived in no other regions than Bavaria and South-Hungary/Northern-Serbia.

Here my results:

Population
North_Sea 31.36%
Atlantic 21.36%
Baltic 11.43%
Eastern_Euro 8.11%
West_Med 12.50%
West_Asian 3.44%
East_Med 9.44%
Red_Sea 0.22%
South_Asian 0.33%
Southeast_Asian -
Siberian 0.60%
Amerindian -
Oceanian 1.14%
Northeast_African 0.07%
Sub-Saharan -

# Population Percent
1 North_Sea 31.75
2 Atlantic 21.63
3 West_Med 12.66
4 Baltic 11.57
5 East_Med 9.56
6 Eastern_Euro 8.21
7 West_Asian 3.48
8 Oceanian 1.15


--------------------------------

Least-squares method.

Using 1 population approximation:
1 French @ 8.595
2 German @ 8.980
3 Dutch @ 11.298
4 Southwest_English @ 11.520
5 Southeast_English @ 12.293
6 Danish @ 13.543
7 Hungarian @ 13.992
8 Norwegian @ 14.166
9 Irish @ 14.383
10 Orcadian @ 14.411
165 iterations.



Using 2 populations approximation:
1 50% North_Italian +50% Swedish @ 4.803
2 50% North_Italian +50% Norwegian @ 5.074
3 50% North_Italian +50% North_Swedish @ 5.788
4 50% Portuguese +50% Swedish @ 5.936
5 50% Spanish_Galicia +50% Swedish @ 6.037
6 50% North_Swedish +50% Portuguese @ 6.384
7 50% Orcadian +50% Serbian @ 6.419
8 50% North_Swedish +50% Spanish_Galicia @ 6.483
9 50% Norwegian +50% Tuscan @ 6.517
10 50% Spanish_Cataluna +50% Swedish @ 6.524
13695 iterations.



Using 3 populations approximation:
1 50% Swedish +25% French +25% Tuscan @ 4.169
2 50% Norwegian +25% French +25% Tuscan @ 4.315
3 50% Norwegian +25% French +25% Greek @ 4.433
4 50% Orcadian +25% Hungarian +25% Tuscan @ 4.546
5 50% Swedish +25% Spanish_Galicia +25% Tuscan @ 4.557
6 50% Swedish +25% French +25% North_Italian @ 4.634
7 50% Swedish +25% Portuguese +25% Tuscan @ 4.737
8 50% Swedish +25% Spanish_Cataluna +25% Tuscan @ 4.760
9 50% Swedish +25% North_Italian +25% Spanish_Galicia @ 4.767
10 50% Norwegian +25% Greek +25% Southwest_English @ 4.783
378909 iterations.



Using 4 populations approximation:
1 French + Swedish + Swedish + Tuscan @ 4.169
2 French + Norwegian + Swedish + Tuscan @ 4.182
3 French + Norwegian + Norwegian + Tuscan @ 4.315
4 French + Greek + Norwegian + Norwegian @ 4.433
5 French + North_Swedish + Norwegian + Tuscan @ 4.482
6 French + North_Swedish + Swedish + Tuscan @ 4.488
7 French + Greek + Norwegian + Orcadian @ 4.502
8 French + Greek + Orcadian + Swedish @ 4.527
9 Hungarian + Orcadian + Orcadian + Tuscan @ 4.546
10 Spanish_Galicia + Swedish + Swedish + Tuscan @ 4.557
11 French + Greek + Norwegian + Swedish @ 4.590
12 French + North_Italian + Swedish + Swedish @ 4.634
13 Norwegian + Spanish_Galicia + Swedish + Tuscan @ 4.650
14 North_Italian + Norwegian + Orcadian + Serbian @ 4.676
15 Hungarian + Norwegian + Orcadian + Tuscan @ 4.690
16 French + Orcadian + Swedish + Tuscan @ 4.729
17 Portuguese + Swedish + Swedish + Tuscan @ 4.737
18 North_Italian + Orcadian + Serbian + Swedish @ 4.739
19 Dutch + French + Swedish + Tuscan @ 4.744
20 French + North_Italian + Norwegian + Swedish @ 4.749

1163521 iterations.

Unknown said...

I was wondering If someone could explain my mothers K13 oracle-4 results? Is the Spanish from Gene flow or is it something more recent?? It does not show up on my results but French Basque does in K15. My mothers is fully English as far as she knows. Father is Scottish

My mothers K13 results:


North_Atlantic 51.09%
Baltic 20.33%
West_Med 15.84%
West_Asian 6.84%
East_Med 3.35%
Red_Sea 0.99%
South_Asian 1.42%
East_Asian -
Siberian -
Amerindian -
Oceanian -
Northeast_African -
Sub-Saharan 0.14%

Using 1 population approximation:
1 Southwest_English @ 3.037
2 Southeast_English @ 4.782
3 West_Scottish @ 5.568
4 Irish @ 5.651
5 Orcadian @ 5.808
6 Danish @ 6.829
7 Dutch @ 7.556
8 German @ 9.060
9 Norwegian @ 11.506
10 French @ 12.278
179 iterations.


Using 4 populations approximation:
1 Irish + Spanish_Cataluna + West_Scottish + West_Scottish @ 2.592
2 Irish + Irish + Spanish_Cataluna + West_Scottish @ 2.594
3 Irish + Irish + Irish + Spanish_Aragon @ 2.608
4 Spanish_Cataluna + West_Scottish + West_Scottish + West_Scottish @ 2.645
5 Irish + Irish + Irish + Spanish_Cataluna @ 2.650
6 Irish + Irish + Spanish_Cantabria + West_Scottish @ 2.686
7 Irish + Irish + Spanish_Castilla_La_Mancha + West_Scottish @ 2.687
8 Irish + Irish + Spanish_Aragon + West_Scottish @ 2.690
9 Irish + Spanish_Cantabria + West_Scottish + West_Scottish @ 2.699
10 French + Southwest_English + Southwest_English + West_Scottish @ 2.708
11 Irish + Irish + Irish + Spanish_Castilla_La_Mancha @ 2.709
12 Irish + Spanish_Castilla_La_Mancha + West_Scottish + West_Scottish @ 2.719
13 French + Southwest_English + West_Scottish + West_Scottish @ 2.725
14 Irish + Irish + Irish + Spanish_Cantabria @ 2.726
15 French + Irish + Southwest_English + Southwest_English @ 2.742
16 French + Irish + Southwest_English + West_Scottish @ 2.763
17 Spanish_Cantabria + West_Scottish + West_Scottish + West_Scottish @ 2.764
18 Spanish_Castilla_La_Mancha + West_Scottish + West_Scottish + West_Scottish @ 2.801
19 Irish + Spanish_Aragon + West_Scottish + West_Scottish @ 2.820
20 French + Irish + Irish + Southwest_English @ 2.851


My K13 results:


# Population Percent
1 North_Atlantic 53.07
2 Baltic 22.52
3 West_Med 12.16
4 West_Asian 9.42
5 South_Asian 2.83


--------------------------------

Least-squares method.

Using 1 population approximation:
1 Irish @ 4.409
2 West_Scottish @ 4.813
3 Orcadian @ 6.203
4 Southwest_English @ 6.534
5 Danish @ 6.570
6 Southeast_English @ 8.044
7 Dutch @ 8.254
8 Norwegian @ 9.523
9 German @ 10.785
10 Swedish @ 11


Using 4 populations approximation:
1 Irish + Irish + Irish + Irish @ 4.409
2 Irish + Irish + Irish + West_Scottish @ 4.466
3 Irish + Irish + West_Scottish + West_Scottish @ 4.553
4 Irish + West_Scottish + West_Scottish + West_Scottish @ 4.670
5 Irish + Irish + Irish + Southwest_English @ 4.726
6 Irish + Irish + Southwest_English + West_Scottish @ 4.811
7 West_Scottish + West_Scottish + West_Scottish + West_Scottish @ 4.813
8 Irish + Irish + Irish + Orcadian @ 4.814
9 Danish + Irish + Irish + Irish @ 4.849
10 Irish + Irish + Orcadian + West_Scottish @ 4.874
11 Danish + Irish + Irish + West_Scottish @ 4.907
12 Irish + Southwest_English + West_Scottish + West_Scottish @ 4.924
13 Irish + Orcadian + West_Scottish + West_Scottish @ 4.963
14 Danish + Irish + West_Scottish + West_Scottish @ 4.993
15 Southwest_English + West_Scottish + West_Scottish + West_Scottish @ 5.062
16 Orcadian + West_Scottish + West_Scottish + West_Scottish @ 5.078
17 Danish + West_Scottish + West_Scottish + West_Scottish @ 5.105
18 Irish + Irish + Irish + Southeast_English @ 5.128
19 Danish + Irish + Irish + Southwest_English @ 5.141
20 Irish + Irish + Orcadian + Southwest_English @ 5.150

X2e1/R1b1b2a1a1 said...

And these are my K15 results:

# Population Percent
1 Atlantic 35.10
2 North_Sea 32.07
3 Eastern_Euro 10.08
4 Baltic 7.57
5 West_Med 7.08
6 West_Asian 6.27
7 South_Asian 1.84


--------------------------------

Least-squares method.

Using 1 population approximation:
1 Southeast_English @ 8.200
2 Irish @ 8.219
3 West_Scottish @ 8.504
4 Danish @ 8.891
5 Southwest_English @ 9.277
6 German @ 11.614
7 Dutch @ 12.166
8 Orcadian @ 12.652
9 French @ 15.478
10 Norwegian @ 15.738

Using 4 populations approximation:
1 Danish + French_Basque + West_Scottish + West_Scottish @ 6.823
2 Danish + Danish + French_Basque + West_Scottish @ 6.840
3 French_Basque + West_Scottish + West_Scottish + West_Scottish @ 6.848
4 Danish + Danish + Danish + French_Basque @ 6.897
5 Danish + French_Basque + Irish + West_Scottish @ 6.947
6 Danish + Danish + French_Basque + Irish @ 6.960
7 French_Basque + Irish + West_Scottish + West_Scottish @ 6.975
8 Danish + French_Basque + Irish + Irish @ 7.090
9 French_Basque + Irish + Irish + West_Scottish @ 7.121
10 French_Basque + Orcadian + West_Scottish + West_Scottish @ 7.127
11 Danish + French_Basque + Orcadian + West_Scottish @ 7.130
12 Danish + Danish + French_Basque + Orcadian @ 7.173
13 Danish + French_Basque + Irish + Orcadian @ 7.213
14 French_Basque + Irish + Orcadian + West_Scottish @ 7.214
15 Danish + French_Basque + Southeast_English + West_Scottish @ 7.265
16 French_Basque + Southeast_English + West_Scottish + West_Scottish @ 7.268
17 French_Basque + Irish + Irish + Irish @ 7.285
18 Danish + Danish + French_Basque + Southeast_English @ 7.301
19 French_Basque + Norwegian + West_Scottish + West_Scottish @ 7.313
20 French_Basque + Irish + Irish + Orcadian @ 7.320

Wondering if the Spanish in my mothers K13 and the Basque in my K15 is Gene flow or something else??thanks

Zachary Shumway said...

My K13 results

Using 4 populations approximation:
1 Danish + Danish + Danish + West_Scottish @ 2.463
2 Danish + Danish + Southeast_English + West_Scottish @ 2.492
3 Danish + Danish + West_Scottish + West_Scottish @ 2.572
4 Danish + Danish + Danish + Southeast_English @ 2.622
5 Danish + Danish + Danish + Danish @ 2.631
6 Danish + Southeast_English + West_Scottish + West_Scottish @ 2.636
7 Danish + Danish + Danish + Southwest_English @ 2.649
8 Danish + Danish + Danish + Irish @ 2.671
9 Danish + Danish + Irish + Southeast_English @ 2.677
10 Danish + Danish + Orcadian + West_Scottish @ 2.712
11 Danish + Danish + Irish + West_Scottish @ 2.731
12 Danish + Danish + Southwest_English + West_Scottish @ 2.762
13 Danish + Danish + Danish + Orcadian @ 2.764
14 Danish + Irish + Southeast_English + West_Scottish @ 2.772
15 Danish + Orcadian + Southeast_English + West_Scottish @ 2.823
16 Danish + Danish + Orcadian + Southeast_English @ 2.840
17 Danish + Southeast_English + Southeast_English + West_Scottish @ 2.871
18 Danish + Danish + Irish + Orcadian @ 2.899
19 Danish + Danish + Irish + Southwest_English @ 2.907
20 Danish + Orcadian + West_Scottish + West_Scottish @ 2.911


Africa9

Using 4 populations approximation:
1 French_Basque + Morocco_Jews + Morocco_Jews + North_Italian @ 1.245
2 French_Basque + French_Basque + Jordanians + North_African_Jews @ 2.795
3 French_Basque + Morocco_Jews + Morocco_Jews + Tuscan @ 2.804
4 Egypt + French_Basque + North_Italian + North_Italian @ 2.965
5 French_Basque + Morocco_Jews + North_African_Jews + North_Italian @ 3.067
6 French_Basque + French_Basque + Jordanians + Morocco_Jews @ 3.293
7 Egypt + French_Basque + French_Basque + North_African_Jews @ 3.409
8 French_Basque + North_African_Jews + Tuscan + Tuscan @ 3.697
9 Egyptans + French_Basque + North_Italian + North_Italian @ 3.731
10 French_Basque + North_African_Jews + North_Italian + Tuscan @ 3.773
11 Morocco_Jews + North_Italian + North_Italian + North_Italian @ 3.776
12 Egypt + French_Basque + North_Italian + Tuscan @ 3.779
13 Egypt + French_Basque + French_Basque + Morocco_Jews @ 3.867
14 Druze + French_Basque + French_Basque + North_African_Jews @ 3.921
15 Bedouin + French_Basque + French_Basque + Morocco_Jews @ 3.951
16 French_Basque + Morocco_Jews + Tuscan + Tuscan @ 3.988
17 Egyptans + French_Basque + French_Basque + North_African_Jews @ 4.035
18 French_Basque + Jordanians + North_Italian + North_Italian @ 4.037
19 French_Basque + French_Basque + North_African_Jews + North_African_Jews @ 4.133
20 Egyptans + French_Basque + French_Basque + Morocco_Jews @ 4.147



*Curious about how this all fits together; what about the African? Any help would be appreciated. Thanks much.

valerio tadici said...

what is considered genuine direct pygmy admixture could be just a remain of early human populations?

valerio tadici said...

because in some GEDmatch calculators i get (very) small percentages of pygmy admixture with no other sub saharan elements.

Melody said...

I've spent hours reading through this thread, trying to self-educate and am lost. I don't know how to interpret the results in most of the tests, and also don't see anyplace to find the explanation of why/how the tests might be used. What would one be looking for in each test ... and then ... how to interpret those results. Please help. In other tests ... FTDNA, Ancestry, 23 and me, I'm coming up with results that don't make sense, based on my genealogy. So confusion about how to interpret the testing here adds to the frustration of not understanding the ethnicities showing up in the other tests. Thank you so much for doing this for all of us, I'm sorry I can't "get it" as I know it would provide me with good answers to the questions about my ethnicity.

Aubrey Potter said...

David - I followed your advice for figuring out Native dna and used the K13. Can you help me figure this out? I am so excited to find out what this native dna is! I was always told that I was German, English and French on my Mom's side and then my father is strangely very different looking from his parents/siblings, pointing toward adoption or something else. I did the 23andme test and sure enough, there's a good chunk 15-20% of something native. Here are my K13 results with my Kit numbering being M154219:

Admix Results (sorted):

# Population Percent
1 North_Atlantic 35.95
2 Baltic 18.33
3 Amerindian 17.55
4 West_Med 11.16
5 West_Asian 7.1
6 East_Med 5.57
7 Sub-Saharan 1.88
8 East_Asian 1.15
9 Siberian 0.94
10 Oceanian 0.37

Single Population Sharing:

# Population (source) Distance
1 West_German 21.41
2 South_Dutch 22.25
3 French 22.72
4 Austrian 23.11
5 Hungarian 23.33
6 North_German 23.64
7 East_German 23.74
8 Southeast_English 24.07
9 Orcadian 24.44
10 North_Dutch 24.67
11 Danish 24.68
12 Irish 24.73
13 Serbian 25.23
14 Southwest_English 25.23
15 West_Scottish 25.24
16 Norwegian 25.37
17 Swedish 25.45
18 Romanian 26.02
19 Portuguese 26.06
20 Spanish_Cataluna 26.32

Mixed Mode Population Sharing:

# Primary Population (source) Secondary Population (source) Distance
1 81% West_German + 19% Pima @ 3.22
2 82.8% West_German + 17.2% Karitiana @ 3.23
3 80.4% West_German + 19.6% Mayan @ 3.25
4 80.4% South_Dutch + 19.6% Pima @ 3.67
5 82.3% South_Dutch + 17.7% Karitiana @ 3.71
6 79.8% South_Dutch + 20.2% Mayan @ 3.77
7 78.9% North_German + 21.1% Mayan @ 5.04
8 78.7% Southeast_English + 21.3% Mayan @ 5.63
9 78.9% West_German + 21.1% North_Amerindian @ 5.64
10 79.7% North_German + 20.3% Pima @ 5.67
11 79.4% Southeast_English + 20.6% Pima @ 5.76
12 78.3% Danish + 21.7% Mayan @ 6.04
13 78.3% North_Dutch + 21.7% Mayan @ 6.05
14 78.3% South_Dutch + 21.7% North_Amerindian @ 6.07
15 81.7% North_German + 18.3% Karitiana @ 6.11
16 81.4% Southeast_English + 18.6% Karitiana @ 6.39
17 78.6% Orcadian + 21.4% Mayan @ 6.44
18 79.1% North_Dutch + 20.9% Pima @ 6.52
19 77% Southeast_English + 23% North_Amerindian @ 6.55
20 79.1% Danish + 20.9% Pima @ 6.59

Suzanne Branch said...
This comment has been removed by the author.
Suzanne Branch said...

Dear Davidski,

Can you tell me how Australian Aboriginal works with these tests and which is the best one to use?

Cody Gearhart said...

hey i have a question. I was wondering if this would tell my ancestry? but more importantly my dads ancestry since i have no idea what he is. i did do geno 2.0 it said italian but i know hes got a lot more than that in him. Thank you for any answer you can give me thank you!

Sámi said...

Hello Davidski!

I have a question. I got the following result in Eurogenes K12:


Population
South Asian 1.10%
Caucasus -
Southwest Asian -
North Amerindian + Arctic 1.28%
Siberian 10.51%
Mediterranean 1.69%
East Asian -
West African -
Volga-Ural 20.76%
South Baltic 17.52%
Western European 20.62%
North Sea 26.53%

How should i interpret this? I know my father was a Saami but what of this is of saami ancestry?

Which calculator is best to use to see actual percentages of saami vs north european heritage?

Ethan Smith said...

Here you can sell and buy both new and used products.
This website is perfect for selling just about anything at all.
More at free classified ads

Spotif Eyeballs said...

David:

I was adopted and I know my birth mother but next to nothing about my biological father. My mother's ancestry is British/Celtic/Scandanavian/German-Dutch. My father was a Castillian Spaniard.

My ancestryDNA came back that I was 50 PER CENT "EUROPEAN JEWISH." My father never spoke of being Jewish. The Spanish family name (which I now claim as my own) does show up in Sephardic Jewish name lists.

Eurogenes K13 4-Ancestors Oracle
Admix Results (sorted):

# Population Percent
1 North_Atlantic 32.39
2 East_Med 18.86
3 Baltic 18.62
4 West_Med 15.00
5 West_Asian 8.19
6 Red_Sea 3.42
7 South_Asian 2.16

Using 1 population approximation:
1 Romanian @ 10.876863
2 Serbian @ 11.039580
3 North_Italian @ 12.910981
4 Bulgarian @ 13.166938
5 French @ 14.519265
6 Tuscan @ 14.732807
7 Austrian @ 14.940455
8 Portuguese @ 16.092104
9 Hungarian @ 16.100468
10 Spanish_Galicia @ 17.198193
11 Spanish_Cataluna @ 17.479113
12 Spanish_Extremadura @ 17.608561
13 Spanish_Murcia @ 18.197008
14 Spanish_Castilla_Y_Leon @ 18.263254
15 Spanish_Valencia @ 18.841171
16 Spanish_Andalucia @ 19.992380
17 German @ 20.211027
18 Spanish_Castilla_La_Mancha @ 20.488518
19 Greek @ 21.220301
20 Spanish_Cantabria @ 21.544460

Using 2 populations approximation:
1 50% Ashkenazi +50% Dutch @ 1.643673


Using 3 populations approximation:
1 50% Ashkenazi +25% Danish +25% Dutch @ 1.483016


Using 4 populations approximation:
1 Ashkenazi + Ashkenazi + Danish + Dutch @ 1.483016
2 Ashkenazi + Ashkenazi + Dutch + Norwegian @ 1.636970
3 Ashkenazi + Ashkenazi + Dutch + Dutch @ 1.643673
4 Hungarian + Lebanese_Druze + Norwegian + Spanish_Valencia @ 1.658654
5 Ashkenazi + Ashkenazi + Danish + Norwegian @ 1.705087
6 Austrian + Lebanese_Druze + Norwegian + Spanish_Valencia @ 1.705304
7 Ashkenazi + Ashkenazi + Norwegian + Southeast_English @ 1.713550
8 Ashkenazi + Ashkenazi + Southeast_English + Swedish @ 1.719311
9 Austrian + Lebanese_Druze + Spanish_Valencia + Swedish @ 1.727489
10 Austrian + Lebanese_Druze + Norwegian + Spanish_Galicia @ 1.757723
11 Austrian + Lebanese_Druze + Norwegian + Spanish_Castilla_Y_Leon @ 1.776700
12 Ashkenazi + Ashkenazi + Danish + Swedish @ 1.788107
13 Ashkenazi + Ashkenazi + Dutch + Irish @ 1.796323
14 Ashkenazi + Ashkenazi + German + Norwegian @ 1.801255
15 Austrian + Lebanese_Druze + Norwegian + Spanish_Cataluna @ 1.802674
16 Lebanese_Druze + Polish + Southeast_English + Spanish_Murcia @ 1.807853
17 Hungarian + Lebanese_Druze + Norwegian + Spanish_Castilla_Y_Leon @ 1.812783
18 Ashkenazi + Ashkenazi + Danish + German @ 1.815230
19 Austrian + Lebanese_Druze + Norwegian + Portuguese @ 1.821689
20 Ashkenazi + Ashkenazi + Dutch + Orcadian @ 1.829508

Jtest Admixture Proportions
Population
SOUTH_BALTIC 9.60%
EAST_EURO 8.52%
NORTH-CENTRAL_EURO 15.38%
ATLANTIC 18.64%
WEST_MED 9.93%
ASHKENAZI 18.48%
EAST_MED 10.67%
WEST_ASIAN 5.71%
MIDDLE_EASTERN 1.64%
SOUTH_ASIAN 1.41%
EAST_AFRICAN -
EAST_ASIAN -
SIBERIAN -
WEST_AFRICAN -


What I am trying to understand is where the "Spanish" part of my ancestry fits in. Some of your tests show various admixtures of Spanish subsets-- but is this being considered as part of my "Ashkenazi" heritage? If 2 ancestors of the K13 4-ancestors Oracle test come back as "Ahekenazhi" then how can the Spanish be part of my Jewish ancestry, since it comes from a third grandparent, which would have to be my mother's side which shows no Spanish anscestry?

I guess I would like to be able to say "I had my DNA tested and my father was a Sepharidc Askenazhi Jew with some Spanish heritage"-- but I am not sure what the DNA is saying.

I am so very confused and I would greatly appreciate your interpretation of my DNA. THANK YOU.

Lauren Gruber said...

Hi David,

Firstly, thanks for your work with this project and facilitating this blog. It's really appreciated! I'm Kit # A657441 and am very new to all this. For background, my most recent heritage is English, German, Italian, Slovakian and some distant French. I'm wondering about my Native American results.

I notice that in the K9 and K10, they appear so small (0.06%), so would this just be noise? The highest percentage I get for Native AmerIndian is 1.05% in K9b. But even in the tests where I do not see a percentage admixture with oracle link, the native american percentages show up in the chromosome. In the chromosome paintings, I usually can see one or two long strains of Native Amerindian DNA. I have had a theory for a while that I'm 1/64 Native American, but have never been absolutely sure. What do you think? Thanks in advance!

K9:

South Asian: 0.97%
Caucasus: 12.48%
Southwest Asian: 4.3%
North AmerIndian: 0.06%
Siberian: 0.78%
Mediterranean: 24.72%
West African: 0.23%
North European: 56.41%

K13:
North Atlantic: 37.23%
Baltic: 24.28%
West_Med: 17.35%
West Asian: 7.94%
East_Med: 7.34%
Red_Sea: 1.92%
South Asian: 1.39%
East Asian: 0.44%
Siberian: --
Amerindian: 0.38%
Oceanian: 0.81%
Northeast Afri: 0.92%
Sub-Saharan:--

Real Deal said...

Hi David,

Newbie here trying to interpret my ancestry with your models. Gedmatch Kit # is A313099. My Father's side is supposedly mostly German with some English/Irish. My Mother's side is mostly Italian with one Polish Great Grandmother. Which tests would be best for me? And if you're able to take a quick glance at my kit #, do the above make sense?

Thanks,
Keith

Donna said...

Hi David,

I am very new to GedMAtch and slowly slowly learning to use the tools. Kit A144875. I have researched my family history at least back to my ggg grandparents. I have taken the autosomal tests and via ancestry.com and transferred the DNA to Gedmatch.

Ancestry.com autosomal results:

Africa < 1%
Mali < 1%

Europe 98%
Great Britain 39% .
Ireland 28% .
Europe West 19% .

Trace Regions 12% .
Iberian Peninsula 4% .
Scandinavia 3% .
Finland/Northwest Russia 3% .
Europe East < 1% .
European Jewish < 1% .

West Asia 1%
Caucasus 1%

If you could plese look at my GedMatch kit number.

Here's my dilemma. The smaller percentages are the ones I am trying to use to get past brick walls in the family history. Ancestry.com tells me I have < 1% African and no native American.

Using some of the Admix tests at GedMatch, I see where I do have some Amerindian and even East Asia.

In my research I have found in some of the family line where Mullato is used with my family line on mothers side for one generation. On my father's side I am trying to find info on gggg grandparents and the only ones that seem to match up are Mullato and black. Also, a reference to this ggg grandfather is that he a dark skinned white man.

Is it possible at the gg and ggg grandparent level that my small percentages of Amerindian and African would show up if these ancestors were possibly Native American or African descent?

Your expertise is greatly appreciated and I hope soon to use your tools skillfully. your input may be the key to my brick wall. thanks so much.

Donna

Davidski said...

Your Native American ancestry, if you do have any that is, might be hiding within the Finland/Northwest Russia 3% in the Ancestry results. Ancestry will probably deny this, but it is possible when dealing with very small levels of admixture. That's because Europeans share genetic ties with Native Americans going back to the Upper Paleolithic, while Finns and Northwest Russians have a bit of recent Siberian ancestry. So your Native American segments might be looking very Finnish to the algorithm at Ancestry.

Keep in mind though, that splitting Eurasian ancestry at such a fine scale is very difficult. What this means is that the minor Amerindian/East Asian admixture you're seeing at GEDmatch might just be noise.

You can try and isolate the potential Native American segments in your genome by looking at a wide variety of the chromosome paintings there. Once you have a good idea where these segments are located, you can see who you match in these regions. You might find that the parts of your genome where Native American ancestry shows up consistently in a wide variety of GEDmatch tests are also the same parts where you share segments with people who have more recently documented Native American ancestry. If so, that would be pretty solid evidence that you do have minor Amerindian admixture.

PaeanApollonos said...

Hi David,
I have a question regarding a couple of the populations that keep popping up in the GEDmatch analyses that I run for my data, specifically for the Baltic and Siberian components.
According to K13, here is the breakdown:

Baltic 3.37%
East_Med 0.32%
South_Asian 4.07%
East_Asian 85.02%
Siberian 3.45%
Amerindian 0.56%
Oceanian 2.89%
Sub-Saharan 0.30%

When I’ve checked via other the data set for EU V2 K15, the result is similar: Baltic 2.24%, Atlantic .89%, Siberian 3.17%.

I guess my question is whether these results are plausible and not an agglomeration of noise. I have noticed that in previous comments anything below about 1% would be considered noise, and the numbers for the Baltic and Siberian components are above that threshold. As a comparison, MDLP K23b postulates 8.23% Tungusic-Altaic (no population data available as of yet, but the Tungusic-Altaic category is exclusive of the Paleo-Siberian, East Siberian, and Ancestral Altaic categories) and Dodecad v3 proposes 2.4% Eastern European with the peak population being the Lithuanians. According to the per chromosome analyses, the information shows up in similar percentages on each chromosome.
I am Filipino, and I know that based on the pedigree I have that we can trace ancestors back to the mid 18th century. I know that a sizeable portion of them immigrated to the Philippines from China. I know that we also have family anecdotes about Spanish ancestry, and that we have a Basque surname that goes back to the beginning of that pedigree. We also have a surname that may very well be Breton or Polish, though I would have no idea about how to determine whether the surname in its present form is merely some orthographic error, or Breton or Polish in actuality (I’ve done some research in both French and Polish and the results are inconclusive). I know that only a very small percentage of DNA from the paper pedigree beyond a couple generations ago will show up in analyses offered by companies like 23andme, and I think that’s the reason why the Basque might not necessarily showing up in the 23andme analysis. The analysis from 23andme generalizes the European portion as being .6%, with .2% being “broadly Southern European” and the .4% being “broadly European”, which I know means that it can be from anywhere else in Europe. Their South Asian estimate is .9% by comparison.

I know that your analyses go further back (I think 23andme says that their data is more applicable to the 500 year range), and I was stumped about the Baltic and Siberian part, especially when considering how those genes made it into my pool. My most recent hypothesis is that a few Balts were captured as slaves and shipped eastward, sometime during the period starting from the Mongolian invasions to the end of the Baltic/Novgorod slave trade at the end of the 17th century. Another possibility that would also account for the Siberian element would be that some of the Balts were sent east when Russia started colonizing Siberia in earnest during the 17th century. The end of either of those hypothetical routes is that the Balt/Siberians were broadly a part of either the Manchu invasions that established the Qing dynasty or the Mongol invasions that established the Yuan dynasty, and that those ancestors intermarried with the Han/Cantonese and emigrated to the Philippines. This seems like the most plausible origin hypothesis for me. The other possibility is that perhaps the hypothetical Polish ancestry from the 1700s is showing up here for whatever reason (although I would think that it would be a much lesser percentage than 3% if it was a single ancestor).

My question is whether or not the percentages of the DNA that is being exhibited by K13 and EU v2 K15 as Baltic and Siberian are merely an anomaly, or whether my hypothesis for their reason for being in my DNA is more recent than it should be (that is, are those DNA percentage sizes from a more distant origin than the 13th century?).

Thank you in advance for your help.

Davidski said...

It seems like you have some minor admixture from North and perhaps West Eurasia, but if so, this is distant, and due to various factors, like ancient links between Eurasians and recombination, the algorithm is having a difficult time classifying these signals.

Baltic might not be the correct classification. It's just what the algorithm is seeing in the data. A higher resolution test, with more markers, or maybe with the use of LD patterns, might be able to give you the correct result.

Real Deal said...

Hi Davidski,

Here are my Eurogenes K13 results:

Population
North_Atlantic 33.70%
Baltic 22.53%
West_Med 16.14%
West_Asian 11.23%
East_Med 10.63%
Red_Sea 1.43%
South_Asian 0.96%
East_Asian 1.49%
Siberian -
Amerindian 0.80%
Oceanian 0.69%
Northeast_African -
Sub-Saharan 0.40%

K9:
Population
South Asian 0.79%
Caucasus 16.15%
Southwest Asian 2.84%
North Amerindian + Arctic 0.35%
Siberian -
Mediterranean 25.73%
East Asian 1.03%
West African 0.42%
North European 52.69%

K36:
Population
Amerindian -
Arabian 0.60%
Armenian 3.00%
Basque 2.34%
Central_African -
Central_Euro 6.46%
East_African -
East_Asian -
East_Balkan 5.75%
East_Central_Asian -
East_Central_Euro 7.89%
East_Med 1.13%
Eastern_Euro 4.38%
Fennoscandian 3.83%
French 9.00%
Iberian 8.85%
Indo-Chinese -
Italian 11.63%
Malayan -
Near_Eastern -
North_African -
North_Atlantic 7.75%
North_Caucasian 7.89%
North_Sea 10.83%
Northeast_African -
Oceanian -
Omotic 0.20%
Pygmy -
Siberian -
South_Asian -
South_Central_Asian 2.33%
South_Chinese -
Volga-Ural 1.91%
West_African -
West_Caucasian -
West_Med 4.18%

My kit # is A313099. My Father's side is supposed to be mostly German with some English/Irish/Dutch here and there. My Mother's side is mostly Italian with one Polish Great Grandmother. I'm really interested in dialing into the German connection. In light of this, do the results above make sense to you? If not, which tests would be best for me? And if you're able to take a quick glance at my kit #, any quick analysis you could provide?

Thanks,
Keith

Jay said...

Newbie here, not sure how to interpret the data on the different test and how accurate which ones are. Oddly the overlap of ancestry.com nails my known family heritage down spot on. I know several lines back to early 1600s.

I know my mothers paternal side is northwest German with a recent mix of Alsace French/Mulhouse left around 1888. By northwest German I mean I know they were married in the same town and church dating back to the 1600s Loningen which is 50km from Dutch border, before that they could have been vikings as Wichmann means Warring People and are several Vikings that have been tried and put to death with that name.

My mothers maternal lineage is a mixed bag, I know German recent, but a long line of French settlers dating back to the 1600s first in Quebec and down the Mississippi to present day MO. This side has a small slice of native american, my seventh great grandmother was full blood. Jean Baptisite Deugire dit LaRose bought his children off the slave owner whom he had children with, this is well documented in the history of Ste. Genevieve, MO. The test on this site has me ranging 2-4% native american. This side was extremely French, the first non French was married into the family in 1908 and after that it has been Germans on this side.

My father is much harder to trace. On his paternal side I know Belgian, but my guess is French/Belgian as my last name is Doré which means golden in French. I know he came to America in 1880 and married into the Whitely family a family a known is Scotch-Irish as I descend from William Whitley a well document family.

My fathers maternal side is 100% German, I mean Germans marrying Germans.

That being said I know my mothers side makeup is mostly a French/German mix, one Englishman and one full blooded Native American my 7th great grandmother.

My father Belgian, German and Scots-Irish, so I am little confused to how to interrupt the different test on you site.

Doing the K13 my results are a little confusing knowing my ancestry. The single population source of Austrian is most confusing, as I know I have a lot of N German and W German. But most surprising is the SE English and Hungarian. Also surprised to see the French to be below Swedish, Danish and Orcadian. The Swedish and Danish I believe would be intermixing of the norther Germans in my background. But the Southeast English make no sense nor does the Hungarian. Any help would be greatly appreciated. With so much known French lineage I am quite surprised it falls below Orcadian/Scottish and Danish.

# Population Percent
1 North_Atlantic 40.99
2 Baltic 27.86
3 West_Med 16.55
4 West_Asian 5.02
5 East_Med 4.77
6 Amerindian 1.93
7 Red_Sea 0.91
8 South_Asian 0.6
9 Sub-Saharan 0.54
10 Northeast_African 0.53
11 Oceanian 0.29

Single Population Sharing:

# Population (source) Distance
1 Austrian 5.91
2 East_German 5.92
3 West_German 6.7
4 South_Dutch 6.81
5 North_German 8.23
6 Southeast_English 9.71
7 North_Dutch 9.95
8 Danish 10
9 Swedish 10.31
10 Hungarian 10.4
11 Orcadian 10.45
12 French 11.09
13 Norwegian 11.12
14 Southwest_English 11.24
15 Irish 11.48
16 West_Scottish 12.32
17 North_Swedish 12.64
18 Croatian 14.97
19 Serbian 15.94
20 South_Polish 16.48