Tuesday, June 26, 2012

My Review of AncestryDNA's Admixture Tool and a Glimpse into the Future of Genetic Genealogy

I have waited to review my experience with my personal AncestryDNA Beta test until I got a good grasp of how the process is working.  So far, I can say that I think there is a lot of promise in AncestryDNA's approach to autosomal DNA analysis, but there are also things that have caused some confusion and concern among the community. I will address just one component of the test today and follow-up with another post to keep this review from becoming unwieldy.

Admixture and the "Scandinavian" Question 

I have been researching the apparent overestimation of the Scandinavian component in AncestryDNA's admixture tool for some time. Like Blaine Bettinger of The Genetic Genealogist, I have been contacted by many people who were surprised by a significant prediction of Scandinavian in their "Genetic Ethnicity" results that is in conflict with their known family history. I have also read the same confusion expressed on many blogs when the testers' results didn't reflect what they know of their family tree. My own results in this area were a bit surprising as well. There are too many examples of this to simply ignore it or to explain it away by non-paternity events (NPEs) or bad genealogies. Blaine Bettinger has already done a thorough job on his recent blog post (Problems with AncestryDNA's Genetic Ethnicity Prediction?) explaining why our admixture results legitimately may not align with our known genealogies, so I won't go too deeply into that at the moment. I highly suggest you read his post for background if you haven't already.

To understand the  perceived problem, let's look at my results as an example:

My "Genetic Ethnicity" Prediction - Click to enlarge

Although I know that approximately 12.5% of my DNA originated from Scandinavia courtesy of my Norwegian great grandmother Fredrikka Herstad, I did not expect the 57% Scandinavian admixture prediction. Of further concern,  I am 25% Finnish on paper and am only predicted as 7% Finnish by AncestryDNA's Genetic Ethnicity feature. For a person who doesn't know much about Finnish DNA, they may assume that my Finnish DNA is simply bleeding into the Scandinavian percentage, however Finnish DNA is quite unique and clusters distinctly separate from the Scandinavian countries (and all others). Although I may have a small amount of Swedish DNA from early migration to Finland, I am quite sure that it is not a significant amount based on my documented genealogy and, especially, my extensive research at both 23andMe and Family Tree DNA where I have myriads of Finnish "cousins" and very, very few Swedish ones. One could argue that these "ethnicity" predictions reach further back than autosomal DNA cousin matching, but I have seen how my Finnish DNA clusters using other admixture tools and am quite confident in my assessment.

Fortunately, there has already been some improvement in AncestryDNA's algorithm as to how it affects my personal admixture results. When I first received my beta results they revealed no British Isles component at all, which definitely gave me pause since my great grandfather George Allen was 100% British. Since then, my percentages have changed, as AncestryDNA warned us they might. My current 28% British Isles is much closer to what I would expect, without taking into account my substantial Colonial American ancestry. I expect their admixture algorithms will continue to improve.

My mother's case is even clearer. Below is her "Genetic Ethnicity" prediction.  Although she is a full 50% Finnish, she is only predicted to possess 23% Finnish DNA. The 62% Scandinavian is especially surprising since my mother has no known Scandinavian ancestry, but she does have significant Colonial American ancestry, predominantly originating from the British Isles, of which her results show none. The Central European component fits well with her known Pennsylvania Dutch great grandmother Ruth Stoalabarger. From this it appears that all of my mother's British Isles and about half of her Finnish is somehow being interpreted as "Scandinavian".

My mother's "Genetic Ethnicity" Prediction - Click to Enlarge

I spoke to Ken Chahine, General Manager for AncestryDNA, on Thursday to try to get an educated perspective on what might be happening rather than jumping to conclusions. Ken was aware of the controversy surrounding the admixture results, but told me that he feels more confident about their predictions since talking to Sir Walter Bodmer and Peter Donnelly of the "People of the British Isles" Wellcome Trust project at the University of Oxford.  He said that they shared information with the AncestryDNA team suggesting that they too are finding a much larger Scandinavian component in the British Isles than expected. Some of their findings can be seen in the recent article here. According to Ken, these discoveries support the notion that the British Isles was a true melting pot long before the United States earned that moniker and further suggests that there was lots of migration between the British Isles and Norway, Sweden and Denmark. Ken feels that these conclusions support AncestryDNA's findings of significant Scandinavian admixture in many who would expect a more substantial British Isles component, like myself. I still have to wonder though if the "Scandinavian" label is a bit of a misnomer. I guess the question is: If the DNA has been in the British Isles for thousands of years, is it still "Scandinavian"? It is impossible to fully address this issue without access to the underlying genetic data on which AncestryDNA's analysis is based.

Further addressing this issue, Ken wanted to remind all of us that our ancestors' DNA can be diluted pretty quickly. Genetic drift comes into play, so we cannot expect our Family Tree to exactly resemble our admixture results. A good explanation of this can be found in Blaine's blog post Everyone Has Two Family Trees - A Genealogical Tree and a Genetic Tree. This concept really only starts to come into play five or more generations deep in our family trees and shouldn't have a substantial effect until we reach deeper than the great great great grandparent level.

Throughout our discussion, Ken emphasized that the exact science of admixture prediction is a "very tough problem to solve". I think we can all agree with him on this point. Anyone who has worked with admixture predictions knows that all of these tools are still rough and will benefit greatly from the increasing availability of reference samples. Ken told me that AncestryDNA now has seven PhDs on staff working on this "problem" - all with computational and mathematics expertise. Especially intriguing is the AncestryDNA team's emphasis on identifying new Ancestry Informative Markers (AIMs) with the goal of ultimately identifying rare alleles (frequency of less than 1% of the population), rather than focusing on long stretches of DNA.

 "Finding these rare alleles will completely transform genetic genealogy." - Ken Chahine

I have to admit that Ken did make me feel a lot better about the direction that AncestryDNA is going with their new autosomal DNA test. He talked at great length about their goal to identify rare alleles from specific areas of the world. Ken's opinion is that the current Illumina chip in use by all three of the companies is not optimal for admixture analysis because the chosen SNPs are too common. According to Ken, for inclusion on the Illumina chips, SNPs are chosen that can be found in at least 5% of the population and are primarily geared toward health-related genes. In contrast, he explained that the real goal should be to find alleles that are RARE and not found in the general population (less than 1%), eventually enabling us to identify these alleles as originating in specific geographic areas and thus pinpointing the ancestral origins of those who possess this marker. (Obviously, full sequencing will provide the eventual solution.)  Ken said his team seeks to use alternative means of discovery (other than chip technology). I was impressed to hear that the AncestryDNA team researches "every single day" for these new AIMs.

Other Points of Note

1. Every day AncestryDNA is updating their algorithms. It is a "living" tool, ever-changing with new input daily. With more plentiful, unique samples and better techniques the admixture and IBD predictions will continually improve.

2. The AncestryDNA team is trying to discover how shared segments relate to where the common ancestor is on our family trees. (I have also wondered if there are any identifiable patterns imbedded in "random" autosomal DNA inheritance.)

3. They are only starting to utilize the Sorenson data now. Their current admixture tool is based on public data.

 4. All data is 100% phased before analysis. That is their first step at AncestryDNA. Ken said that it never occurred to him not to phase the data and that he was surprised that the other companies were not doing this as well. He explained that the computational requirements are enormous due to the phasing and that they looked at three or four different phasing engines - and found that all were roughly comparable.

The Future

There are still no answers regarding whether/when AncestryDNA will be providing the underlying genetic data to their customers, i.e. - specific matching segments or downloadable raw data, but Ken assured me that they are working on it. He said that they are still in the decision-making process and are currently evaluating the best way to deliver this data. He strongly dispelled the rumors that I had heard that AncestryDNA has decided that they will not provide the raw data to customers. It is indeed essential that this happens because blind faith has no place in a scientific environment such as this. Both Family Tree DNA and 23andMe release the genetic data, in line with their stated belief that all personal genetic information is the property of the individual, thus allowing for outside analysis and intellectual challenge. How can anyone disagree with this?

This conversation with Ken was an exhilarating one in regard to what is in store for us genetic genealogists. He possesses a promising vision for the future of genetic genealogy. The potential for strides in genetic genealogy is mind-blowing with a company pouring these kinds of resources into its advancement and working toward a well-defined, singular goal. This begs certain important questions: What will happen to the competitors if they are not able to match this level of financial commitment? Will AncestryDNA commit to an "open access" model that enables the citizen scientist to share in the new discoveries made, following in the footsteps of Family Tree DNA and 23andMe? Will the increased competition in this sector lead to leaps forward for all? Time will tell, but I can guarantee that the coming months will be very exciting ones in the world of genetic genealogy.

[10/26/12: This test is now out of Beta, so you can order it here.]
[5/19/13: I have noticed that many people who are looking for information about the AncestrybyDNA test are coming here. This is not the same company or test. For more on this, please read this post.]

Disclosure: I received both of the tests discussed above complimentary from AncestryDNA during their early beta testing phase.

96 comments:

  1. Very interesting! I will be looking forward to seeing what they do with providing the raw data and showing specific segments where people overlap. Thanks for another insightful post!

    ReplyDelete
  2. Dear CeCe,
    Thanks for this nice summary. I would like to point out that genetic drift really doesn't appear to be a major issue in autosomal DNA biogeographical analysis (BGA) until you get back to about the great-great-great-great grandparent level. At that point you would expect that on average each great-great-great-great grandparent would be contributing 110.9 cMs of autosomal DNA for people tested by 23andMe and 105.6 cMs of autosomal DNA for people tested by FTDNA's Family Finder. At this point and further back in one's family tree genetic drift could indeed be an issue. However, for people such as yourself whose ancestry is 1/4 Finnish, your BGA at Ancestry.com should relatively closely reflect this 25% Finnish contribution or Ancestry.com's BGA accuracy must be called into question. Genetic drift should not be used an excuse for inaccuracy in this situation. It would be preferable if Ancestry.com didn't call the origin of specific segments of DNA if they can't be highly confident that those segments are from a specific ancestral population. Hopefully, Ancestry.com's BGA accuracy will continue to improve with time. The whole issue of accurate BGA is indeed a complicated one, particularly when one is only using data from the current SNP arrays for the analysis. When we move to complete genome sequencing we will see an incremental improvement in the accuracy of BGAs.
    Sincerely,
    Tim Janzen

    ReplyDelete
    Replies
    1. Hi Tim,
      Thanks for this important clarification regarding genetic drift and your usual insightful comments. I added a sentence to the relevant paragraph in this regard so as not to mislead anyone.
      CeCe

      Delete
  3. CeCe, Many thanks for this interesting report. Do you think you might be able to ask Ancestry at some point if they could publish a list of the reference populations they are using? Presumably they have incorporated data from the People of the British Isles Project as some of these sequences have now been made publicly available. The Ancestry results are somewhat meaningless without this information.

    ReplyDelete
    Replies
    1. Hi Debbie,
      I will certainly put that question on my list for any potential future conversations with the folks from AncestryDNA. (There were just so many that I should have asked!) When was the data from the People of the British Isles Project made public?
      Thank you,
      CeCe

      Delete
    2. CeCe, Some of the sequences from the People of the British Isles Project are now included in the 1000 Genomes Project.

      Delete
  4. What if your full sibling has one of those rare AIMs which are geographically significant and you don't? Both of you come from the same ancestors. I have 7 full siblings, so I guess I have 7X a chance to find rare AIMs in them not found in me. Genealogical tree and genetic tree differences within siblings.

    ReplyDelete
    Replies
    1. Hi Ponto,
      I'm sure that will occasionally happen since we all inherit our DNA in unique patterns from our ancestors. We have already seen that some siblings inherit African or Asian segments in their Ancestry Paintings while others do not. It is the luck of the draw when we are talking about inheritance from distant ancestors.
      Thanks!
      CeCe

      Delete
  5. CeCe,

    Great post. I am glad to read that AncestryDNA will give access to the raw data at some point.

    Regards,

    Larry

    ReplyDelete
    Replies
    1. Thanks, Larry. I certainly hope that they will eventually do so. CeCe

      Delete
    2. CeCe,

      For whatever it is worth, I was told on the phone last week by one of Ancestry's DNA reps that they WILL be releasing the raw data in the future.

      Delete
    3. Jan, Thanks for sharing that. I'm glad to hear that is the rep's impression too.
      CeCe

      Delete
  6. Thanks for another perspective on AncestryDNA questions. I will echo Debbie that a definition of reference populations would be a nice touch. To test AncestryDNA use for my personal research I asked the same research question for AncestryDNA and Family Tree DNA and blogged about my experience. While I am no expert, it was interesting to use both labs for an autosomal research strategy as they stand now.

    Vagueness in scientific testing will beget assumptions and confusion. So I appreciate your report of your interview as it was very informative for the future of AncestryDNA

    ReplyDelete
    Replies
    1. Thanks for reading and for the comment, Heather. Can you please post a link to your relevant posts, so we can also benefit from your insights?
      CeCe

      Delete
    2. Genetic Tales of My Father: An Autosomal DNA Strategy: http://geneabrarian.blogspot.com/2012/06/genetic-tales-of-my-father-autosomal.html

      Delete
    3. Ah yes, you are the Geneabrarian. Great post! I read it during my research on this subject. Thanks for the link. :-)

      Delete
  7. I mistakenly made a post on Blaine's Blog thinking it was here. Me bad. This new information you provided on the phasing aspect got me thinking. If Ancestry is using phased data this could explain a lot. As I said in my post on Blaine's blog. My Swedish Norwegian ancestors are the most recent immigrants and are 25% of my tree. (One grandparent 100% Swedish/Norwegian) Whereas other 3 grandparents ancestry is a lot more mixed and all colonial. So the segments from my Scandinavian grandparent would more clearly paint "intact." Whereas my others would not be as clearly defined. If the data is phased this might skew the results in favor of the clearly Scandinavian segments. Whereas the English/Scottish/German mixes would be more scrambled. Using GEDMATCH ancestry painting utilities this skews as well---as sometimes ---like on my "African" segments it shows 100% or 75% African when that is not the case. Its just if you have that "identifying" rare value it calls for "African" when in fact it is African from one parent and European from the other.

    From my Y-DNA project and the Y signatures in my tree I know that not all my British is of "Viking" origin. So something is clearly amiss. I can now verify the DNA of 7 of my 16 great-great-grandparents. That is: previously unknown individuals at the 3-4th cousin levels match me and our trees match---so there's no unknown Scandinavian hiding in the closet on those lines. At least not in the past 350 years and likely another 700 years before that.

    My guess is that population specific markers pull the results toward them more strongly than those markers which are more generic. The evidence of 3 major ancient pools of British DNA: Celtic, Norse, and Germanic should not place someone with clearly centuries old roots in Britain to being called for as Norse.

    My colonial English and more recent Scottish is at least 50% of my ancestry yet I show NO English at all. My matches confirm that my DNA still carries this 50% so clearly something does not add up.

    Here's a comparison:
    Best Rough guestimate
    1/2 English/Scottish (via Ireland)
    1/4 Swedish/Norwegian
    3/16 German/Austrian/Swiss/French
    1/16 Mixed Roma/Portuguese/African/Turkish/Paupan/Native American

    Ancestry.com
    60% Scandinavian (shows Sweden, Norway, Denmark on their map)
    31% Central European (shows France, Germany, Belgium)
    9% Unknown

    So what I would have expected is something more like this:
    25% Scandinavian
    50% British
    20% Central European
    5% Other

    Even accounting for the Vikings it would still be no more than 35% Scandinavian. 60% just doesn't add up.

    I will send this feedback to Ancestry as well. And I look forward to their refinements. I do love their product even absent RAW data and matching segments which I do hope are coming. As always thanks for your insight and help.

    ReplyDelete
    Replies
    1. Hi Kelly,
      This is a very interesting analysis and echoes many of my thoughts. Thank you for posting it for further consideration of this issue.
      I agree with you that their product has some very great aspects to it, which I will cover in my review - Part II.
      CeCe

      Delete
    2. I have the opposite "problem".....not any Scandinavian showing up for me, when my father tests at 53%, and 47% British Isles. He does id as my parent.....
      My results are supposedly 93% British Isles, and 7% Central European. Oh, and my Mom....she came in at 80% Central European, 16% British Isles, and 4% unknown. From a 21,000 person tree, I figure I should be 50% CE, 25% BI, and 25% Scandinavian.

      Delete
    3. Same here, Glen. I have heavily documented Norwegian ancestry, yet not one percent showed up on Ancestry's test. I did, however, rake in 25% Central European, which was news to my mother, who has done genealogy for over forty years. From whence did THAT come from, and where's the Gilbertsons?

      Delete
  8. As an African American I could care less about white nationality not being from an exact country.The main thing I care about with Ancestry.com's Ancestrydna test is that it doesnt miscalculate my African ancestry and give me "Asian" percentages like 23andme does whether or not I really have Asian or Native American ancestry.

    ReplyDelete
    Replies
    1. Hi Anthony, May I ask what your AncestryDNA breakdown is? Aren't you at all interested in from where your European ancestry originates?
      Thanks for your comment,
      CeCe

      Delete
    2. yes,I am interested in all my ancestry.I have never done Ancestrydna from Ancestry.com or any other admixture test but my dad and maternal grandparents have done the outdated AncestryBYdna 2.5 from the DNA Diagnostics Center.My dad was given "32 European and 68%African", my granddad was given "27% European and 73% African",and my grandma was given "21% European and 74% African and 5% East Asian" which we are not sure is noise or if it is really Indigenous American assigned to East Asian(the confidence intervals for East Asian and Indigenous American on my grandma's results were really high,above 10,but we don't know if that really means anything). But I would love to do Ancestry.com's Ancestrydna especially after seeing actor Blair Underwood on NBC's WhoDoYouThinkYouAre? and seeing his African breakdown which is something that no other test has done.Sorry if I may have offended anyone by my above post,and speaking of Blair Underwood, that is something to think about how his European was "20% French/Swiss and 6% German",I wonder how many French swiss and Germans were in Virginia dealing with Africans then. Maybe Ancestry.com does have an issue with their European countries,but as an African American,I am alittle more interested in an African breakdown like I saw with Blair Underwood(whether or not that's all that accurate either) and Ancestrydna calculating my African as African and not some "Asian" like 23andme gives just about all their black test takers,whether African Americans or African from Africa, because they only have one reference population.

      Delete
    3. No, no offense taken here, Anthony. Unfortunately, AncrestryDNA is not yet offering the detailed African breakdown to the public that we all saw on WDYTYA for Blair Underwood. Those detailed percentages appear to have been arrived at using reference populations in Sorenson's database. However, according to Ken, they have not used the Sorenson reference pops as of yet in their results for the public. I am confident that they will do so in the future, but since they aren't yet, your belief in their admixture superiority for African American customers may be misplaced until such time as they incorporate the Sorenson database into their algorithms.
      Thanks for your comment and best of luck with your testing goals once you move forward with testing. :-)
      CeCe

      Delete
  9. I'm one of those shocked people! I was expecting a higher percentage of Central European since I know on paper, as far back as my 5th great grandparents at least that they came from Germany and Bohemia. However, it was the smallest percentage at 5% and instead I was 35% Scandinavian with only two known ancestors of that in my tree and not until I get to about 12 generations back on my father's side.

    ReplyDelete
    Replies
    1. Laura, you are definitely not alone! I have read examples of this all over the Internet, which is why I started looking into it. I'm sure all of us will be watching closely for adjustments to the algorithms, particularly throughout the Beta phase, as well as the upcoming publications from the People of the British Isles Project.
      Thanks for your comment,
      CeCe

      Delete
    2. I, too, am shocked by my breakdown. My mother's mother came from Sweden. Her father came from Norway. I grew up in a home that kept the Swedish traditions at Christmas and my mother and grandmother prepared many Swedish dishes. I have traced both sides and there is no doubt they were in Scandinavia back as far as I can trace them through parish and household records. My father was mostly of German and Dutch ancestry with some British as well. I am absolutely certain of the German and Dutch ancestry. But, guess what. My Ancestry DNA prediction is 95% British Isles and 5% uncertain. I simply don't believe it. It is as if they erased my mother. I know I wasn't adopted. Half of my genes come from my mother. Both of her parents came from Scandinavia. I even have a 98% match who turned out to be descended from my Swedish grandmother's first cousin. So, it is impossible for me to believe the ethnicity prediction is accurate. I do think the genetic matching is accurate. It has matched me up with many who have Scandinavian and/or European with no British at all. I have a lot of matches with the shared surname, Schenck. Schenck, is a Dutch and German name. I have connected with many of them in my tree. When I ask Ancestry about all of this they tell me to wait and the results should change over time. Other than that, they ignore my concerns. But they haven't changed over time. Not even a little. I feel like I have been cheated out of my money. I am tired of hearing people make excuses for them.

      Delete
  10. I have not found a lot of information and your readers seem to be very well informed. I have the completely boring results of 91% British Isles and 9% Unknown. Most of my family is England/Scotland/Ireland, but has really been in the US since Colonial times. I did not expect any surprises. The only exception is that I have 1 grandfather who was full blooded French (25% to me). Is French ancestry mixed with British Isles or should that be Southern European and maybe is included in the 9% unknown?

    ReplyDelete
    Replies
    1. Hi ~lynnie,
      Yes, I have exceedingly intelligent readers! :-)
      It is difficult to say why the French isn't showing up in your admixture. France is included in AncestryDNA's "Central European" label, so I would certainly think that you should have a percentage of CE. (Within Europe it is difficult to identify unique DNA signatures due to lots of migration, which is presumably why France is lumped in with Germany, Austria, Belgium, Netherlands, Switzerland, Slovenia, Czech Republic, Luxembourg, and Liechtenstein under this CE label.) Keep watching for changes. Hopefully, your results will become more in line with your known ancestry as AncestryDNA progresses with their research.
      Thanks for reading!
      CeCe

      Delete
    2. My AncestryDNA results were similar to what Lynnie describes in her post (06/27/12). My results were 97% British Isles and 3% unknown, while I know that my grandmother was 100% French.
      There were many "4th Cousin" member matches listed as having 96% confidence, but only 3 actually appear to have any real match.
      The Genetic Ethnicity page shows a pin map that appears to show locations taken from my online tree, most of which are in France so I'm not really clear on the connection between the pin map and my AncestryDNA results.
      Tom

      Delete
    3. My AncestryDNA results were similar to what Lynnie describes in her post (06/27/12). My results were 97% British Isles and 3% unknown, while I know that my grandmother was 100% French.
      There were many "4th Cousin" member matches listed as having 96% confidence, but only 3 actually appear to have any real match.
      The Genetic Ethnicity page shows a pin map that appears to show locations taken from my online tree, most of which are in France so I'm not really clear on the connection between the pin map and my AncestryDNA results.
      Tom

      Delete
    4. My AncestryDNA results were similar to what Lynnie describes in her post (06/27/12). My results were 97% British Isles and 3% unknown, while I know that my grandmother was 100% French.
      There were many "4th Cousin" member matches listed as having 96% confidence, but only 3 actually appear to have any real match.
      The Genetic Ethnicity page shows a pin map that appears to show locations taken from my online tree, most of which are in France so I'm not really clear on the connection between the pin map and my AncestryDNA results.
      Tom

      Delete
  11. Thanks for your efforts and the update CeCe!

    ReplyDelete
  12. I don't know if anyone else ask this, did they say anything about accepting for a fee the raw data from 23andme or Family Tree for admixture analysis?

    ReplyDelete
    Replies
    1. Hi Socrates,
      No, they did not. I have asked them before and I know they are considering it, but no word on whether it will happen.
      Thanks,
      CeCe

      Delete
  13. I dont get it. As far as I understand most unique SNP's is found at the individual level in their haplotypes. You dont need to look for single SNP's with that or this low frequencty as it seems mutations already have happend in haplotypes consisting of common available SNP's. I suspect that the reason why certain SNP's appears polymorphic is because they have a larger mutation rate than other SNP's that have remained unchanged since the Neandertals.

    Looking at single SNP's that dominate in one population or area take to much resources when you can identify easily mutated SNP's in haplotypes constructed from common SNP's. As far as I can tell they are doing the wrong strategy to solve the problem, but I am not a expert but an amateur, I may have not done the math.

    Of course then you have to rely on the phasing of genotypes to haplotypes using software with unrelated individuals but the error rate should as I understand be low.

    I have primarly worked to understand the genetics of Fennoscandia, but I suspect it would be much easier to find f.ex those Native American specific haplotypes than exclusive mutations from f.ex the Saami.

    ReplyDelete
  14. RE: "Ken's opinion is that the current Illumina chip in use by all three of the companies is not optimal for admixture analysis because the chosen SNPs are too common. According to Ken, for inclusion on the Illumina chips, SNPs are chosen that can be found in at least 5% of the population and are primarily geared toward health-related genes. In contrast, he explained that the real goal should be to find alleles that are RARE and not found in the general population (less than 1%), eventually enabling us to identify these alleles as originating in specific geographic areas and thus pinpointing the ancestral origins of those who possess this marker."

    A question I have is this: Will AncestryDNA absolutely guarantee customer that any of the Autosomal DNA SNPs being analysed have significance SOLELY to Genetic Genealogy and NONE for Genetic Health / Medical Reasons.

    The POBI - People of the British Isles scientists which AncestryDNA is now talking with are funded by Welcome Trust - a Pharmaceutical Company.

    Plus, A new Ancestry.com Inc. Board Member, Dr. Paul R. Billings is a board certified Clinical Geneticist and the Chief Medical Officer at Life Technologies Corporation.

    CeCe ... I do think this a big deal as some of us do want nothing to do with any company which will share our Private DNA analysis for Health and Medical reasons within the testing company or later on to 3rd parties.

    ReplyDelete
    Replies
    1. George, I agree that this is a big deal. I have wondered the same thing, especially since their release form says that they won't be using it for health-related testing at this time, but appears to leave the option open for later.
      Thanks for your comment,
      CeCe

      Delete
    2. CeCe,

      The following is the AncestryDNA OLD and NEW T&Cs (Terms and Conditions). Their NEW ones went into effect 6-3-12 ... and YES they can modify them anytime. Which to me is not very consumer friendly at all and as my earlier post demonstrates ... they are looking at other Profit Centers distinct from Genetic Genealogy.

      There are some keywords the management and their attorneys intentional selected for their intent as a "Legal or Contractual Out" for AncestryDNA to use RIGHT NOW the DNA results of a group (1 or more persons) any way they wish.

      One keyword is INDIVIDUAL (1 person) as noted below in both their old and new versions. "... and not for INDIVIDUAL medical or diagnostic purposes."

      So if AncestryDNA used your DNA results and my DNA results (a GROUP of 2) they can do anything they wish with it for monetary gain such as putting a patent on it or some other device of ownership.

      Also, I hope an attorney will read this posting and clarify other T&Cs from 23andme / AncestryDNA / FTDNA which are not consumer friendly such as agreeing to arbitration in any dispute and agreeing not to participate in any Class Action lawsuits.

      IMHO, it's going to take an ISOGG or some other consumer friendly Genetic Genealogy organization to effectively advocate for changes within these companies (and 3rd parties such as GedMatch.com) voluntarily or getting a bill passed in Congress doing so.

      This issue will blow wide open once we have a sub $1000 DTC Full Genome test available.

      To me, FTDNA is the white knight in this debate and do all they can not to analyze anything which may have health or medical implications.
      These T&Cs only apply to U.S. customers .... those in Europe / EU and other locations may have more stringent guidelines than we do in the U.S. I wonder what Canada and other coutries are doing?

      New Ancestry.com DNA LLC T&Cs (eff. 6-3-12 for existing users): http://dna.ancestry.com/legal/termsAndConditions.aspx
      4. DNA Testing (paragraph 1 - sentence 1)
      "All DNA testing performed by AncestryDNA on samples submitted for testing or by uploading a digital version of a DNA analysis is done for genealogical research only, including population and ethnic group-related analyses, and not for INDIVIDUAL medical or diagnostic purposes."
      (paragraph 1 - last sentence)
      " In addition, you understand that by providing any DNA to us, you acquire no rights in any research or commercial products that may be developed by AncestryDNA that may relate to or otherwise embody your DNA."

      Old Ancestry.com DNA LLC T&Cs: http://dna.ancestry.com/legal/termsAndConditionsArchive.aspx
      4. DNA Testing (paragraph 1 - sentence 1)
      "All DNA testing performed by AncestryDNA on samples submitted for testing or by uploading a digital version of a DNA analysis is done for genealogical research only, including population and ethnic group-related analyses, and not for INDIVIDUAL medical or diagnostic purposes."
      (paragraph 1 - last sentence)
      " In addition, you understand that by providing any DNA to us, you acquire no rights in any research or commercial products that may be developed by AncestryDNA that may relate to or otherwise embody your DNA."

      Delete
  15. Could somebody please explain to me how they go about "phasing" our data? Is this just used for the ethnicity predictions or is it used in determining cousin relationships or both? I must be confused because I thought the only way to phase our data was to have a parent tested as well.

    Also was wondering if anybody else who has a subscription is still waiting for an invitation to test? I think I signed up on the first or second day we could sign up to be notified and never have been offered a test and I have an ancestry subscription.

    ReplyDelete
    Replies
    1. Dean - I also had signed up very early on. UPDATE - I got my "invitation" two days ago (with a time window - $99 offer expires after 7 days). So, hopefully yours has arrived too.

      I was surprised how long it took to arrive - they must have a lot of people on the waiting list.

      With 3/4 of my paper ancestry as full Swedish, I will be very surprised if it isn't dominated by "Scandinavian".

      Eric

      Delete
  16. Dear Anders,
    I think you are missing the point about low frequency autosomal SNPs. These low frequency SNPs could be quite helpful for pinpointing the geographic region that an ancestor was from and in some cases could even be linked to specific ancestors. However, we are not yet at the point that we can start studying low frequency SNPs in great detail. For now, we need to concentrate on doing chromosome mapping using phased data. As complete genome sequencing becomes much more commonplace then we can start mapping all of the low frequency autosomal SNPs and other autosomal variants. Autosomal SNPs or variants that originated in a specific ancestor 500 or fewer years ago will be particularly helpful for genetic genealogists once they have been correctly linked to the original ancestor.
    Sincerely,
    Tim Janzen

    ReplyDelete
    Replies
    1. Yes I understand your point Tim. These SNP's would be like unique events like defining haplogroups on the Y-Chromosome. However finding these exclusive SNP's most likely need complete genomes to be effectivly found. Affordable complete genomes is still some years ahead for most of us.

      Delete
  17. Dear Dean,
    There are several ways you can do phasing. You can phase data from close relatives using two parent/one child trios. David Pike and I have written programs that do that. There are also programs such as Beagle that phase data without having the benefit of data from close relatives. See http://faculty.washington.edu/browning/beagle/beagle.html. Those programs use data from reference populations. The programs impute phasing by comparing small portions of the autosomal SNP genotype of interest with the reference populations. This type of phasing can be done reasonably accurately, but it is computationally challenging to do this type of phasing. Results are generally more accurate if you have close relatives such as one's parents available for comparison when you phase the data. About 94% of one's autosomal SNP data can be accurately phased if you have the data for both parents available for comparison. The other 6% could be relatively accurately phased using a program such as Beagle and large reference samples.
    Sincerely,
    Tim Janzen

    ReplyDelete
    Replies
    1. I have done phasing of 289k SNP genome data as a routine now for many months with BEAGLE using larger sets of individuals within 10-30 minutes on a standard desktop. I dont see much problem to scale it up to include the whole 550k or 1M SNP without using so extremly much more time. The problem for me have so far beeing handling the output as some massive output files are generated.

      Delete
    2. If we are to believe the reconstructed haplotypes made by phasing software like BEAGLE mutations happends all the time using common SNP's. I see it when investigating interesting haplotypes in detail. You can have a haplotype of common SNP's: ACGTACGTACGT and then a another haplotype ACGTGCGTACGT. Obviously this haplotype have mutated. Population structure software like Chromopainter exploits this kind of haplotype information. I can using it f.ex see that the number of counted mutations is less vs closer populations than more distant ones.

      Delete
  18. Dean, I am also still waiting for my invite. I signed up after CeCe posted the link here which was before the public announcement on 3 May 2012. I am an ancestry subscriber (I think you have to be to sign up?).

    ReplyDelete
  19. Thanks for this article. I just submitted my sample and was doing some extra reading about the process. As a scientist I appreciate the detailed information as well as the comments by your very informed readers.

    ReplyDelete
  20. CeCe, this is really encouraging to hear what is being planned for the future in genetic genealogy.

    I used to adhere to the school of thought that says if it looks like a duck it's a duck. In other words I assumed there should be correlation between phenotype and genotype. Thus, noting your physical appearance in your profile photo, 57% Scandinavian didn't seem unreasonable.

    Now I know that such thinking is naive, at least in the present generation of BGA tests. My older brother is about 5 inches taller than me, blond and blue. I had (mostly grey now) dark brown hair, green eyes, and tanned much better than my older brother. He has had hundreds of liquid nitrogen removed skin cancers, which I've never had despite never having used sunscreen and being frequently out in the sun kayaking on lakes and the ocean.

    When I had my brother tested at 23andme (we're both on V3), I assumed he would have a few percent more northern components, and a similar amount lower Mediterranean than I had in Dodecad. So it was a surprise when the results came in! At Gedmatch I'm M121954 and my brother is M152440. It turns out he is 13.14% less northern European than I am (East European and West European combined), and 5.33% more Mediterranean than I am. He's even higher on South Asian than I am .59% versus .15%.

    ReplyDelete
  21. I forgot to add that I'm Dave123 at 23andme.

    ReplyDelete
  22. I would like to clear up some confusion in the posts around the 28th June. The Wellcome Trust has nothing whatsoever to do with the pharmaceutical industry.

    In 1995 Glaxo acquired / merged with Wellcome to become Glaxo Wellcome; then merged with SmithKline Beecham to become GlaxoSmithKline - and now GSK. In 1995 Wellcome Holdings sold all of its remaining investments and shares in Wellcome, the pharmaceutical company - and obtained about £425 million pounds for them.

    They then set up the Wellcome Trust, which is now the largest funder of basic research in the UK. It has an enviable world-wide reputation and high ethical standards.

    Glad we cleared up the confusion over the name Wellcome - and what it means in this context. So the People of the British Isles has no connection to the pharmaceutical industry. It is the stated aim of the PoBI Project to make its information freely available - so the pharmaceutical industry can access that the same as we can. It is worth remembering too that the University of Oxford, in a different Department, is part of the 1000 Genomes project and can certainly access information and data which is confidential and we cannot - and I think that also carries over into the PoBI Project.

    ReplyDelete
    Replies
    1. Thanks so much, Brian for the much needed clarification.
      CeCe

      Delete
  23. After a hike on a nearby mountain I was pleasantly surprised to see Ancestry.com's offer to participate in the DNA Beta test on my inbox. I purchased immediately. Since I'm 3/4 British Isles and 1/4 German, in this early version it will probably come out mostly Scandinavian until improvements are introduced.

    Dave Schroeder (Dave123 at 23andme)

    ReplyDelete
  24. Hello CeCe
    I enjoy reading your posts.

    After two testing kits and many months of waiting...my Ancestry DNA results show that I'm:
    58% Central European
    19% Scandinavian
    15% British Isles
    6% Eastern European
    2% Uncertain (23andMe shows me at 2.1% Neanderthal - maybe that's Ancestry's way of avoiding the topic).

    I was a bit surprised because I know my documented family history and it contains zero evidence of Scandinavian ancestry. To be honest, I had already accepted that my mother's German and Swiss Mennonite lines would, at a minimum, place me in the 50% range for Central/Eastern European ancestry. I can't get Scandinavia out of my head. My logical mind is churning through all that I've learned about Scandinavian conquests into present day Germany and Scotland (amongst other locations) and I'm trying to make the puzzle pieces fit again. I'm sure that it will, one day.

    Now I know how people feel on Dr. Gates' program and WDYTYA as they find out that what they know to be true isn't as true as they once thought. Like others, I'm looking forward to future testing. We've come a long way from the first 12 marker tests. In our lifetime, we should see the inexpensive full genome test and supporting tools for analytical research that will allow us to trace our genetic ancestry in all directions.

    ReplyDelete
  25. Hello,
    Thank you for addressing the "Scandinavian" question. I have this problem in reverse. My Ancestry Autosomal results are 56% British Isles, 28% Scandiavian, & 16% Finnish/Volga-Ural. Two grandparents (Mother's parents) were from Sweden and I have traced their families back into the 1600's in the Skåne region of Sweden. I have relatives there that have confrmed my work on these lines. My father's Mother was from the Sachsen region of Germany and I have traced that lie into the 1500's. My father's father's family were German but resided near Danzig, West Prussia. I have no clue where the 56% British Isles could come from. My only possiblility seems to be from the Danzig ancestors, which I have only been able to trace to the 1700's. I have read that many people of Scottish descendent immigrated to Danzig when it was part of the Hanseatic League. I am also confused why my results show zero Germanic (Central Europe) origins.

    ReplyDelete
  26. I received my results today. I thought I would be 100% German. I have all German surnames going back to the 1500s in some cases; documented ancestors in German-speaking Switzerland; and all over Germany. My father's family was listed as ethnic Germans in Hungary before they immigrated to US in early 1900s.

    I was very surprised to receive results of 90% British Isles; 10 % Persian/Turkish. The British Isles part is just doesn't make sense. I know the Germans invaded England but did the English migrate to Germany? Of course, my German ancestors came from multiple locations throughout Germany......anyone have any thoughts?

    I

    ReplyDelete
  27. This comment has been removed by the author.

    ReplyDelete
  28. CeCe - Thanks for an interesting article. I have one fairly off-the-wall thought on the high percentage of Scandinavian in so many British trees. My Y-DNA haplogroup [actually my male cousin's] is I1,I2B but my ancestors for that line are German Swiss. We tested through a FamilyTree DNA Surname Project and the I1 group is predominately Scandinavian However the breakdown showed a fair percentage of the same group branching into England and Ireland with a minor sweep into Germany/Switzerland. If they are calling the whole haplogroup "Scandinavian" it could explain some of the confusion.

    ReplyDelete
  29. Thank you very much for this analysis and information. I'll admit that I'm a novice when it comes to understanding genetic information, but I am very surprised by the results that my dad received today from his test. I have extremely well documented information that suggests he should be at least 75% Scandinavian - 3 of his 4 grandparents were 100% Swedish, going back at least 200-300 years. The 4th grandparent (his mother's mother) is more of a mystery, but she likely has roots in England and other parts of western Europe. She could have some Scandinavian heritage as well, which is why I say that my dad should have at least 75%. The results we received were as follows:

    Scandinavian 46%
    British Isles 34%
    Central European 16%
    Uncertain 4%

    Unless I'm misunderstanding how this works, either the test is just way off, or there is a major skeleton in the family closet.

    I sent this feedback to Ancestry, but I'd love to hear any insight that others have! Thanks!

    ReplyDelete
  30. As far as the last few hundred years, I know of zero Scandinavian relations. My mother was born in Croatia and her family before her and my father's family from the British Isle, yet my results were 50% Scand; 25% BI and 25% E.Europe. My concern is this seems to be a snapshot in time but it is too far back..maybe a 1000 years or more..to be of any real use. Jaxsun

    ReplyDelete
  31. As I've mentioned before, I predicted 100% German. I have a few lines that go back to the 1400s but my results showed 90% British Isles/10% Persian/Turkish! Ancestry responded to my inquiry. Here's part of their message:

    your DNA is compared to our database of samples from people with verified heritage. We have collected an extensive database from all over the world but we obviously can't have samples from everyone (and not everyone has a verifiable heritage). So it's likely that your DNA just isn't matching any of the individuals that we have as samples of German ancestry and that it doesn't match our German samples. the British Isles must be the next closest match (since we all share some DNA in common).

    2. Right now this new DNA test is still in BETA and we are developing the database of known samples. I know that we will continue to grow the number of verified heritage samples for all the regions of the world and this should make matching more accurate. While this sample is definitely your sample, in the future as we grow the database, it's possible that your results can change. For example, if we are able to obtain more samples of verified heritage from Germany it's possible that you will be a match for them, and even a closer match that you currently are for the British Isles portion of your result.

    Jaxsun - I have the same concern that these results are all too far back in time - I hope that it will someday make sense to me.

    I've been British for two weeks:) Still getting used to it!

    Chris

    ReplyDelete
  32. FYI. If you tested with ancestryDNA and you no longer subscribe to ancestry.com, the monthly fee for "Ancestry Connections" is $34.95.

    I believe Ancestry Connections allows you to continue to receive matches, the ability to contact your matches and to view posted gedcoms.

    AncestryDNA may not be cost effective for some people.

    The one time fee for the autosomal tests of FTDNA ($289) & 23andMe ($299) includes continuous matches AND your raw data.

    It is important to note that currently, AncestryDNA does not allow you to download your raw data.

    With the initial $99 fee for AncestryDNA and 6 months of Ancestry Connections ($209.70), you have already surpassed the costs of their competitors.

    Something to think about before ordering!

    ReplyDelete
  33. This comment has been removed by the author.

    ReplyDelete
  34. I have a different, yet similar problem. My father's mother was half English. We have traced our genealogy back to the Mayflower and historical figures in the Revolutionary War. It is all very substantiated in family bibles, published family histories and genealogical records. So imagine my surprise when my Ancestry.com Autosomal DNA test found absolutely NO British Isles in my ethnic breakdown with no Scandinavian either. I came up as 29% Middle European (some of that is my grandmother's French and German side). I admit that I am a rather complicated ethnic mix (my mother was full blooded Sicilian, hence the Southern European at 24% and Eastern European at 28%, from the Greeks who settled in Sicily), some Turkish and some Middle Eastern (from my one Jewish Grandfather) but I would have at least expected to see around 10-12% from the British Isles. Perhaps they are just not that good with getting the British Isles right yet in general?

    Thanks for letting us know that things may improve in the future. When I called Ancestry.com the agent told me that perhaps my father didn't pass down any of his English heritage to me. My gut instinct is telling me that this is wrong and it's the inadequacy of their testing that isn't finding it.

    On a personal level, this has been very disappointing. After being elated to find out that my family was so involved in such important aspects of early American history, it hurts to in effect be told that I do not share in their heritage. In some ways I wish I had not taken the test!

    ReplyDelete
    Replies
    1. Hi Renee, I am so sorry to hear this. Please do not be discouraged. You very likely DO share in this heritage. Just look through your match list and I'll bet you find matches who descend from some of your early American ancestors. When you find that, you will know that you did inherit DNA from them.

      As far as AncestryDNA"s "ethnic" breakdown, I think it has a long way to go and it will improve over time. Honestly, I don't think any of these detailed percentage breakdowns are very accurate right now. That may change with the new Geno 2.0 and 23andMe's new Ancestry Painting. Time will tell.

      British Isles does seem especially difficult to pin down with these tools. When I first tested at AncestryDNA, I had NO British Isles either and I have an immigrant great grandfather who was FULL British (should contribute about 12.5% of DNA) plus a lot of early Colonial NE ancestors. After awhile, my percentages changed and I now have 28%! This feature is still in Beta, so please take it with a grain of salt. I think it was wrong of the customer service agent to tell you that. People should not be reassessing their genealogies or self-identities based on this test - at least not yet.

      Thanks for your important comment.

      Delete
    2. Thank you for posting this. I also contacted Ancestry about my results and was told that the results were extremely accurate. This was after I had explained the significant distress my results (which would appear to completely exclude one parent) had caused in our family. I asked if I could purchase another test to try to clear up what was going on with the results, and that request was flat out denied. As a long-time Ancestry.com subscriber, I was really disappointed in the lack of understanding about how upsetting these results can be when they are vastly different from what is expected.

      Delete
    3. Emily, That is awful and, obviously, upsetting. I would suggest you test at either 23andMe or Family Tree DNA in order to get your raw data which can be analyzed by third party "citizen scientists". If you test at either of those places, I can look at your results and help figure out if there really is a discrepancy between your family tree and your DNA.
      Please send me an email, so we can discuss the details privately.
      yourgeneticgenealogist@gmail.com

      Delete
    4. Genetic Genealogist, thanks for your reply. Unfortunately none of the matches Ancestry has given me (which are supposedly 96% accurate to be 4th - 6th cousins) show up any names in my tree whatsoever save for one, which I will get to later. And I am looking very carefully through the entire list of names in their trees. There are a few people with British Isles in them but their ancestors come from the places like Philadelphia and the South and I have absolutely no link to the South unless you go back to the 1500s and figure that one person might have gone to Virginia instead of New England. But how remote is that? Certainly not 4th-6th Cousin.

      The one tree that had a name match was to a woman named Betsy Daniels, who was supposedly born in Oneida NY in 1822. The person's tree did not have any parents but there were children, although with a different last name, going to the present all with different last names every generation (I guess this was all through a female side). Now my Daniels family lived in Vermont, which is not that far away, so there is a chance that this person is related to me, but I see the relationship as more remote than 6th cousin, and even if it is that, it's not very significant given how long ago the common link might be (at least seven generations ago or more).

      I also seem to be matched to some people with Jewish ancestry, but again no names match at all. And they are being called a certain percentage of "European Jewish", which I did not get. They are being matched with me on who knows what - some supposedly 8% Middle Eastern. The funny thing is that some of these people are coming up as matched to me under other things, like Southern European. Makes no sense.

      And what about the 11% Turkish? If there is any Turkish in me it must have been many generations ago. I am intimately acquainted with my genealogy going back to at least the 1700s on all sides save for perhaps the Jewish side, and there are no Turks in there whatsoever.

      Even those who were supposedly matched with me on the Italian are not showing up in my tree, and I am fully half Sicilian with documented heritage going back solidly in ONE TOWN in Southern Sicily until at least the early 1700's! (Thanks to a tireless Cousin who went there and thoroughly researched it).

      So I agree with you that Ancestry.com has a long way to go with this and that it can only improve with time. I presume as the database increases, so will the accuracy. Thanks for reconfirming that for me.

      Delete
    5. After closer inspection, I'm being matched mostly to people of "European Jewish" descent (even though I didn't get that in my results) in the "Central European" section. Go figure. The Daniels match was also in that section, and that should really be associated with "British Isles", but I didn't get that result either.

      I found one more possible name match that Ancestry.com didn't pick up. It was one of those European Jewish people who had a woman in their tree born in 1911 with the last name "Lampert". My Jewish Great Grandmother's maiden name was "Lampard" or "Lampart". These people lived in Queens, NY - My Jewish great grandparents settled in Brooklyn. So that's a possible match. She has no information on the woman's origins or parents.

      Delete
  35. When I received my results with 90% British Isles from my two German parents, I also called and asked for another test in the hopes that I could test another family member and get it right. Ancestry told me that my 90% results were "very strong" and I could count on them. I was also told I should expect some changes in the future. And no additional test....

    So far, none of my numbers have changed at Ancestry but I feel more certain that I'm not adopted (only half joking - the thought crossed my mind) since I tested with 23andme. There I have much more of a heritage mixture and so far a large portion of my matches are with people of German ancestry. I recommend trying another test. BTW there is a discount available right now through giltcity.com for $225 and a further discount of 3 for $200/ea. I'm signing up more family members:)

    I hope you take some comfort in the fact that this is a burgeoning science and as it is perfected we will only get more and more information in the future.

    ReplyDelete
    Replies
    1. My understanding is that autosomal cousin matching, regarding of phasing etc., "taps out" at around 10 generations due to the nature of random recombination. And considering that most family tree researchers can go back about 10 generations with existing paper trails, I think the it is TOTAL HYPE that the future of automosal testing be dubbed exciting. As a serious researcher, I am starting to question what value autosomal testing can offer... other than verification of existing family trees. Now that alone is of value... but not quite the excitement being toted...

      Delete
  36. I don't know why 23 and me is one of the favorites. They are horrible! They have not updated populations since 2008 and are extremely behind the times in capturing native american. They are completely inept at answering questions because they simply don't know what they are doing. I would not recommend them to anyone but for the medical info. DNA consultants is the best out there to date and way ahead of everyone else. Dr. Yates has performed some of the only accurate and inclusive studies on native american and melungeon dna and now everyone is trying to catch up with his data.

    ReplyDelete
    Replies
    1. Jan - I couldn't agree with you less. I do not recommend DNA Consultant to my readers. 23andMe has been conservative with their admixture predictions, but they are updating their Ancestry Painting in the next few weeks and it shows a lot of promise.
      Thanks for your comment,
      CeCe

      Delete
    2. Hi CeCe. Why in the world would you not recommend DNA Consultants? I would love to hear the reasoning. After over 10 yrs of working on genealogy and DNA results/tests/ companies, they are by far the best available with regards to pulling the info together and the only one that works specifically on NA Dna- the old views are peeling away and I have found he is usually months ahead of everyone else. The already have the Rare alleles test. His results is the only one that lines up with the paper trail that I have unearthed- even the things I did not know were there...

      Delete
    3. I posted something last night but I don't know that it made it through. At any rate---to repeat- I would love to know your reasoning on this- but now I get it....I read the front page of your blog- and you are an "ambassador" for 23andMe. After over 10 yrs working with genealogy and DNA tests/companies etc. While you certainly are entitled to your opinion- this is your site after all!.... 23andme- just now adding populations? They are completely behind the times. Conservative is one thing-I find them to be way beyond conservative to just plain out-of-date. They are clueless on Native American. DNA consultants results have paired up completely with my paper trail genealogy over time. The results have fallen right in line even on things I did not know about until digging deeper into the paper trail of genealogy. There is no way I would ever recommend them to anyone and several other folks I am in contact with via Family tree and Ancestry.com have had the same experience. Hopefully Family Tree DNA (autosomal) will also come around as they are quite limited as well. The results with DNA Consultants and Tribes are so specific they use them for criminal cases -profiling issues on certain matters and have actually led to solving cases. They can not do that with the others companies as of yet. I appreciate what you are doing....I just wholly, but respectfully disagree.

      Delete
    4. That's fine, Jan, We are all entitled to our educated opinions. However, I must point out that my status as a volunteer advisor to 23andMe in no way affects my opinions as reflected in my blog. As I have said many times, I am an equal opportunity criticizer and praiser reagrding all of the major companies in this space. I have demonstrated that time and time again.
      As far as DNA Consultants - I do not have personal experience with them, but if they do such terrific work, then why is it that none of the very experienced genetic genealogists who are active in the community recommend them?
      Thanks for your comments.
      CeCe

      Delete
  37. After reading all of the above posts, I would like to make a comment. There is no 100% anything. Ever. Unless your grandfather's eternal lineage from the dawn of time dates back to England --from the inception of his first relative. Not possible. Even if that were the case- doesn't mean you will get that particular shake out from the alphabet soup of DNA. It may show up in a sibling, and not you. Additionally, I think folks get confused about location vs actual DNA-lineage. For example, my 5th Ggrandmother -Mary Jemison. (The Scottish adopted Seneca) --her parents were born in Ireland. They are NOT Irish. There parents lineage dates back to Highland and Lowland Scots. The DNA is Scottish- where they were born is Ireland. It makes them Irish by birth, however, their DNA and actual originating bloodline is Scottish along with some Dutch and French thrown in. Being born somewhere does not make you of that DNA- it makes you of that heritage which is completely different. because you or your parents were born in a certain place, does not mean that's what will show up in your DNA, it may show up as modern day populous match, but that can be different than your actual bloodline ancestry.
    One more example- my adoptive father- his parents where born in Hungary- but they were not of Hungarian DNA, they were transplanted Germans.

    ReplyDelete
  38. My DNA results on Ancestry.com just came in.
    59% Scandinavian
    22% Southern European
    12% Persian/Turkish/Caucasus
    7% Uncertain.
    No Native American, so... My Aunt Dora was telling stories, as I suspected. Interesting.

    The concentration of Scandinavian DNA was a real surprise to me. The Southern European seems to confirm what we have suspected on one of the Maternal names. The balance was news to me.

    ReplyDelete
  39. I just got my ancestry results back today & I was very surprised to see 97% British Isles & 3% Uncertain. None of my French ancestry seems to have shown up at all although in looks that's the part of the family I resemble the most.

    What is most confusing to me though is I am matched to someone who they think is probably 4th cousin. Her results show Central European 72%. Eastern European 28%. So what does that mean? That I'm somehow matching with her in that mysterious 3% that is uncertain?

    I'm so confused. Was hoping you could explain that to me. Thanks!

    ReplyDelete
  40. I feel my "ethnicy results" are Wayyyy off. DNAancesty.com shows my "ethnicty" as 50% Scandinavian(can't be true), 47% Central European(that's about right,) and, 3%undetermined. I feel that because of my many, many English ancestors I should show 47% British Isles; 47% Central European; 3% Scandinavian and 3% undetermined. It seems the "cousin matching" results are working out well. I've "matched" with quite a few people. However, I match to others through their and my English Ancestors "in common." Soo...If my "ethnicity results" show 0% British Isles. Why am I "matching" so many of the other's British ancestors??? It doesn't make sense. Are all the 13 "documented" and approved by the Mayflower Society, Mayflower Pilgrims who are ALL my great. great....gr... grandparents, John Alden, Priscilla Mullins, Myles Standish, Isaac Allerton, Stephen Hopkins, Henry Samson; Francis Cooke; Richard Warren, ect., all Scandinavian? Also, my paternal grandmother was 1/2 English and 1/2 Scottish. Now I might concede that the Scot might be Scandinavian, but they married into English Royalty.. so is all of the English Royalty really Scandinavian? I don't know. As far as the "matching cousins part", many family trees that I "match" don't have enough people on their family tree to be able to match. I can be pretty sure they are cousins, but can't "prove" it. Unless they work on their family tree I won't be able to find an "ancestor in common." I am wondering why these people with only 28 people on their family tree would have gotten a free DNA test from ancestry.com?

    ReplyDelete
  41. Very interesting post, thank you Cece. I would like to offer my thoughts as well. I am french, born in France to Northern French parents. Expect for a small branch from the south, most of my family is Flemish.

    My results were strange to me. (Dnaancestry)

    77 per British isles
    14 per Southern European
    9 percent. Persian.??

    My family tree is well documented going back in some cases 1000 years....my conclusion is that the French DNA is not well documented in their database yet

    ReplyDelete
  42. My French and german did not show. I showed eastern European that I did not know I had. Does ancestry.com have enough data base to be accurate? Does it test both maternal and paternal sides or should I get my brother to test too? Seems like 23 and me is more informative about ancestry and maybe I should test with them instead any suggestions?

    ReplyDelete
  43. Very interesting reading today. I too was shocked to see I was 92% Scandinavian, 2% uncertain and 6% Eastern European. My mother is full German on both sides with them immigrating up the Mississippi to Illinois in the 1850's. Could they have ancestors way back who came from Scandinavia? I guess. My dad's side is all English/German/little Scottish and Irish. Does that mean that the test relates back to the Vikings thus the 92% Scandinavian?
    I have a gift for a AncestrybyDNA test. Can a female take this test or is better for a direct male relative of the line take it if I am trying to prove or disprove my great grandfather was a illegitimate? Thanks for your reply.

    ReplyDelete
  44. Thanks for your report it was very helpful. I don’t know too much about my ancestries, I had my DNA tested by Ancestry DNA and found out the results was 72 percent Scandinavian 20 percent west Africa and 8 percent unsure. I know my mother’s father was Cajun and my mother’s mother had Africa blood I don’t know how much, and my father was Scottish. It seem that everyone tested by them have a percent of Scandinavian. My mother’s father was Cajun I don’t see anything about this, unless this was the 8 percent uncertain. I'm new to this DNA testing and I'm my mother's only child. Just wanted to find out where my famly came from.

    ReplyDelete
  45. Thank you, CeCe, for a great article and posters for your insight. I stumbled upon your article from a link on the 23andMe page. I had my DNA tested by AncestryDNA. It came up 78% British Isles, 15% Scandinavian, 7% Other, which is pretty much what I had expected. I was adopted and found my mother's family a few years ago. They were colonial Americans of English descent. I know nothing about my father. I was told I was part Norwegian. There are no Norwegians in my mother's family tree. So I wonder if the 15% Scandinavian is from my father's side or the result of a Viking raid on the English coast. Likewise, I am unable to determine how much of the 78% British Isles is English. Certainly all of my mother's ancestors going back generations are of English ancestry but having been born in a Catholic facility, I suspect there is Irish on the other side. Would a 23andMe test help narrow this down?

    ReplyDelete
    Replies
    1. Hi Pat,
      Yes, it would definitely help you learn more about your ancestral origins to test at 23andMe. Don't be surprised if that Scandinavian percentage disappears or is greatly reduced. Most people who have tested at both companies find this to be the case. I believe that AncestryDNA is overestimating Scandinavian for most people.
      Thanks for your comment!
      CeCe

      Delete
  46. Thanks CeCe! I was looking into doing the test to put me closer to finding my biological father and/or his relatives, since the sight promises to connect users to other relatives. What are your thoughts on my plan of action? do you think the test will help? Any feedback is greatly appreciated.

    ReplyDelete
    Replies
    1. Hi Shalante,
      AncestryDNA would not be my first choice for an adoptee just starting their search, although adoptees have found their bio family there. I would suggest testing at 23andMe first ($99) because they give more data and tools. It is best to be in all three databases if at all possible. However, there are other important non-DNA avenues of adoption search as well. Have you already exhausted those? Please send me an email: yourgeneticgenealogist@gmail.com so we can discuss further or join one of the groups linked on the side of my blog - AdoptionDNA and/or DNA Newbie for assistance in your search.
      I'm glad that you wrote!
      CeCe

      Delete
  47. How does the DNA testing work when you know nothing about your father?

    ReplyDelete
  48. Cece,

    I have a curious situation that I don't know what to make of. I had my husband and his father both tested with Ancestry. My father-in-law is 53% British Isles, and yet his son is 0%. I understand if I were comparing two different databases, but how is this possible that he isn't at least 3% British? All of the other populations match. I haven't tested the mother-in-law, but the other percentages are certainly possible. I just don't understand how the same database would calculate the same genes differently.

    Any ideas?

    Thanks,
    Ashley

    ReplyDelete
  49. I wonder about the test being able to evaluate Jewish ancestry, is this type of testing able to show weather a person is of Sephardic or Ashkenazi descent? I was born in a country that I never felt the connection with except for some outer aspects of that country which I remember to like about. Deep, deep inside I always felt I never belonged there. As a child my father told me something about my family and perhaps showed me something, but I was too little to remember the extend of the conversation, only vague pieces still in my mind like the fact that I know some information which I cannot remember. Through out my life I have the desire to retrieve that information and even considered regressive therapy but not sure if that is the path to follow in order to retrieve what I know but cant remember to the full extend. I had a very strange dream as a child as well but you will have to be on my shoes to believe it. I know I want to go home, but how can I prove my ancestry related to that particular group when many important details are missing from the equation? Only pieces here and there and what the mind has forgotten. In an attempt to know since I have very little resources to go by, I decided to find out in a very controversial way, and I do not know if that way can be a reliable one, but I am so curious to know and find out. I came across some information about the pendulum to inquire information, and came across some answer which appear to confirm the link with that group. I am new to that and it is strange. Does anyone has any comments about finding Jewish ancestry though DNA testing with Ancestry.com? Please share. Thanks!

    ReplyDelete
  50. CeCe,
    Great post. I have been considering doing a DNA profile on myself for a while now. It has long been known that I have some native American ancestry but recent evidence from a family tree an uncle had been working shows we may have two different tribes from two very different parts of the country. Will this test be able to differentiate between the two?

    A second set of questions, which is more of a concern, is what does ancestry.com do with the sample? Is it destroyed? Or banked? Do any of the tests/companies allow you to stay anonymous?

    ReplyDelete
  51. CeCe,
    Great post, thank you for the information. I do have a question with what ancestry.com does with the sample. Do they destroy it? Or is it banked for eternity? As much as I want to know my profile it makes me a little nervous having my DNA out there. Do any of the companies do it anonymously? Thank you!

    ReplyDelete
    Replies
    1. Sorry I missed this before. Any of the companies will destroy your sample at your request. Family Tree DNA allows testing under aliases and, I imagine, AncestryDNA does too. 23andMe doesn't encourage it, but since this is a DTC test, you are always free to test under a different name.

      Delete
  52. I had an ancestry test done. It gave interesting results. I am British, and I got a map showing my origins localised to different regions. It showed a lot Russian and some Polish, Bavarian and Norwegian among others. As for the British Isles, it showed Scottish, English and Irish. My father has found Scandinavian, Irish and Gypsy ancestors through genealogy and my grandmother (on my mother's side) said she had danish ancestry, but I couldn't work out where a lot of the other locations came from.

    ReplyDelete