Chromosome Browser War

There has been a lot of discussion lately, and I mean REALLY a lot, about chromosome browsers, the need or lack thereof, why, and what the information really means.

For the old timers in the field, we know the story, the reasons, and the backstory, but a lot of people don’t.  Not only are they only getting pieces of the puzzle, they’re confused about why there even is a puzzle.  I’ve been receiving very basic questions about this topic, so I thought I’d write an article about chromosome browsers, what they do for us, why we need them, how we use them and the three vendors, 23andMe, Ancestry and Family Tree DNA, who offer autosomal DNA products that provide a participant matching data base.

The Autosomal Goal

Autosomal DNA, which tests the part of your DNA that recombines between parents every generation, is utilized in genetic genealogy to do a couple of things.

  1. To confirm your connection to a specific ancestor through matches to other descendants.
  2. To break down genealogy brick walls.
  3. Determine ethnicity percentages which is not the topic of this article.

The same methodology is used for items 1 and 2.

In essence, to confirm that you share a common ancestor with someone, you need to either:

  1. Be a close relative – meaning you tested your mother and/or father and you match as expected. Or, you tested another known relative, like a first cousin, for example, and you also match as expected. These known relationships and matches become important in confirming or eliminating other matches and in mapping your own chromosomes to specific ancestors.
  2. A triangulated match to at least two others who share the same distant ancestor. This happens when you match other people whose tree indicates that you share a common ancestor, but they are not previously known to you as family.

Triangulation is the only way you can prove that you do indeed share a common ancestor with someone not previously identified as family.

In essence, triangulation is the process by which you match people who match you genetically with common ancestors through their pedigree charts.  I wrote about the process in this article “Triangulation for Autosomal DNA.”

To prove that you share a common ancestor with another individual, the DNA of  three proven descendants of that common ancestor must match at the same location.  I should add a little * to this and the small print would say, “ on relatively large segments.”  That little * is rather controversial, and we’ll talk about that in a little bit.  This leads us to the next step, which is if you’re a fourth person, and you match all three of those other people on that same segment, then you too share that common ancestor.  This is the process by which adoptees and those who are searching for the identity of a parent work through their matches to work forward in time from common ancestors to, hopefully, identify candidates for individuals who could be their parents.

Why do we need to do this?  Isn’t just matching our DNA and seeing a common ancestor in a pedigree chart with one person enough?  No, it isn’t.  I recently wrote about a situation where I had a match with someone and discovered that even though we didn’t know it, and still don’t know exactly how, we unquestionably share two different ancestral lines.

When you look at someone’s pedigree chart, you may see immediately that you share more than one ancestral line.  Your shared DNA could come from either line, both lines, or neither line – meaning from an unidentified common ancestor.  In genealogy parlance, those are known as brick walls!

Blaine Bettinger wrote about this scenario in his now classic article, “Everyone Has Two Family Trees – A Genealogical Tree and a Genetic Tree.”

Proving a Match

The only way to prove that you actually do share a genealogy relative with someone that is not a known family member is to triangulate.  This means searching other matches with the same ancestral surname, preferably finding someone with the same proven ancestral tree, and confirming that the three of you not only share matching DNA, but all three share the same matching DNA segments.  This means that you share the same ancestor.

Triangulation itself is a two-step process followed by a third step of mapping your own DNA so that you know where various segments came from.  The first two triangulation steps are discovering that you match other people on a common segment(s) and then determining if the matches also match each other on those same segments.

Both Family Tree DNA and 23andMe, as vendors have provided ways to do most of this.  www.gedmatch.com and www.dnagedcom.com both augment the vendor offerings.  Ancestry provides no tools of this type – which is, of course, what has precipitated the chromosome browser war.

Let’s look at how the vendors products work in actual practice.

Family Tree DNA

1. Chromosome browser – do they match you?

Family Tree DNA makes it easy to see who you match in common with someone else in their matching tool, by utilizing the ICW crossed X icon.

chromosome browser war13

In the above example, I am seeing who I match in common with my mother.  Sure enough, our three known cousins are the closest matches, shown below.

chromosome browser war14

You can then push up to 5 individuals through to the chromosome browser to see where they match the participant.

The following chromosome browser is an example of a 4 person match showing up on the Family Tree DNA chromosome browser.

This example shows known cousins matching.  But this is exactly the same scenario you’re looking for when you are matching previously unknown cousins – the exact same technique.

In this example, I am the participant, so these matches are matches to me and my chromosome is the background chromosome displayed.  I have switched from my mother’s side to known cousins on my father’s side.

chromosome browser war1

The chromosome browser shows that these three cousins all match the person whose chromosomes are being shown (me, in this case), but it doesn’t tell you if they also match each other.  With known cousins, it’s very unlikely (in my case) that someone would match me from my mother’s side, and someone from my father’s side, but when you’re working with unknown cousins, it’s certainly possible.  If your parents are from the same core population, like Germans or an endogamous population, you may well have people who match you on both sides of your family.  Simply put, you can’t assume they don’t.

It’s also possible that the match is a genuine genealogical match, but you don’t happen to match on the exact same segments, so the ancestor can’t yet be confirmed until more cousins sharing that same ancestral line are found who do match, and it’s possible that some segments could be IBS, identical by state, meaning matches by chance, especially small segments, below the match threshold.

2. Matrix – do they match each other?

Family Tree DNA also provides a tool called the Matrix where you can see if all of the people who match on the same segment, also match each other at some place on their DNA.

chromosome browser war2

The Matrix tool measures the same level of DNA as the default chromosome browser, so in the situation I’m using for an example, there is no issue.  However, if you drop the threshold of the match level, you may well, and in this case, you will, find matches well below the match threshold.  They are shown as matches because they have at least one segment above the match threshold.  If you don’t have at least one segment above the threshold, you’ll never see these smaller matches.  Just to show you what I mean, this is the same four people, above, with the threshold lowered to 1cM.  All those little confetti pieces of color are smaller matches.

chromosome browser war3

At Family Tree DNA, the match threshold is about 7cM.  Each of the vendors has a different threshold and a different way of calculating that threshold.

The only reason I mention this is because if you DON’T match with someone on the matrix, but you also show matches at smaller segments, understand that matrix is not reporting on those, so matrix matches are not negative proof, only positive indications – when you do match, both on the chromosome browser and utilizing the matrix tool.

What you do know at this point is that these individuals all match you on the same segments, and that they match each other someplace on their chromosomes, but what you don’t know is if they match each other on the same locations where they match you.

If you are lucky and your matches are cousins or experienced genetic genealogists and are willing to take a look at their accounts, they can tell you if they match the other people on the same segments where they match you – but that’s the only way to know unless they are willing to download their raw data file to GedMatch.  At GedMatch, you can adjust the match thresholds to any level you wish and you can compare one-to-one kits to see where any two kits who have provided you with their kit number match each other.

3. Downloading data – mapping your chromosome.

The “download to Excel” function at Family Tree DNA, located just above the chromosome browser graphic, on the left, provides you with the matching data of the individuals shown on the chromosome browser with their actual segment data shown. (The download button on the right downloads all of your matches, not just the ones shown in the browser comparison.)

The spreadsheet below shows the downloaded data for these four individuals.  You can see on chromosome 15 (yellow) there are three distinct segments that match (pink, yellow and blue,) which is exactly what is reflected on the graphic browser as well.

chromosome browser war4

On the spreadsheet below, I’ve highlighted, in red, the segments which appeared on the original chromosome browser – so these are only the matches at or over the match threshold.

chromosome browser war5

As you can see, there are 13 in total.

Smaller Segments

Up to this point, the process I’ve shared is widely accepted as the gold standard.

In the genetic genealogy community, there are very divergent opinions on how to treat segments below the match threshold, or below even 10cM.  Some people “throw them away,” in essence, disregard them entirely.  Before we look at a real life example, let’s talk about the challenges with small segments.

When smaller segments match, along with larger segments, I don’t delete them, throw them away, or disregard them.  I believe that they are tools and each one carries a message for us.  Those messages can be one of four things.

  1. This is a valid IBD, meaning identical by descent, match where the segment has been passed from one specific ancestor to all of the people who match and can be utilized as such.
  2. This is an IBS match, meaning identical by state, and is called that because we can’t yet identify the common ancestor, but there is one. So this is actually IBD but we can’t yet identify it as such. With more matches, we may well be able to identify it as IBD, but if we throw it away, we never get that chance. As larger data bases and more sophisticated software become available, these matches will fall into place.
  3. This is an IBS match that is a false match, meaning the DNA segments that we receive from our father and mother just happen to align in a way that matches another person. Generally these are relatively easy to determine because the people you match won’t match each other. You also won’t tend to match other people with the same ancestral line, so they will tend to look like lone outliers on your match spreadsheets, but not always.
  4. This is an IBS match that is population based. These are much more difficult to determine, because this is a segment that is found widely in a population. The key to determining these pileup areas, as discussed in the Ancestry article about their new phasing technique, if that you will find this same segment matching different proven lineages. This is the reason that Ancestry has implemented phasing – to identify and remove these match regions from your matches. Ancestry provided a graphic of my pileup areas, although they did not identify for me where on my chromosomes these pileup regions occurred. I do have some idea however, because I’ve found a couple of areas where I have matches from my mother’s side of the family from different ancestors – so these areas must be IBS on a population level. That does not, however, make them completely irrelevant.

genome pileups

The challenge, and problem, is where to make the cutoff when you’re eliminating match areas based on phased data.  For example, I lost all of my Acadian matches at Ancestry.  Of course, you would expect an endogamous population to share lots of the same DNA – and there are a huge number of Acadian descendants today – they are in fact a “population,” but those matches are (were) still useful to me.

I utilize Acadian matches from Family Tree DNA and 23andMe to label that part of my chromosome “Acadian” even if I can’t track it to a specific Acadian ancestor, yet.  I do know from which of my mother’s ancestors it originated, her great-grandfather, who is her Acadian ancestor.  Knowing that much is useful as well.

The same challenge exists for other endogamous groups – people with Jewish, Mennonite/Brethren/Amish, Native American and African American heritage searching for their mixed race roots arising from slavery.  In fact, I’d go so far as to say that this problem exists for anyone looking for ancestors beyond the 5th or 6th generation, because segments inherited from those ancestors, if there are any, will probably be small and fall below the generally accepted match thresholds.  The only way you will be able to find them, today, is the unlikely event that there is one larger segments, and it leads you on a search, like the case with Sarah Hickerson.

I want to be very clear – if you’re looking for only “sure thing” segments – then the larger the matching segment, the better the odds that it’s a sure thing, a positive, indisputable, noncontroversial match.  However, if you’re looking for ancestors in the distant past, in the 5th or 6th generation or further, you’re not likely to find sure thing matches and you’ll have to work with smaller segments. It’s certainly preferable and easier to work with large matches, but it’s not always possible.

In the Ralph and Coop paper, The Geography of Recent Genetic Ancestry Across Europe, they indicated that people who matched on segments of 10cM or larger were more likely to have a common ancestor with in the past 500 years.  Blocks of 4cM or larger were estimated to be from populations from 500-1500 years ago.  However, we also know that there are indeed sticky segments that get passed intact from generation to generation, and also that some segments don’t get divided in a generation, they simply disappear and aren’t passed on at all.  I wrote about this in my article titled, Generational Inheritance.

Another paper by Durand et al, Reducing pervasive false positive identical-by-descent segments detected by large-scale pedigree analysis, showed that 67% of the 2-4cM segments were false positives.  Conversely, that also means that 33% of the 2-4cM segments were legitimate IBD segments.

Part of the disagreement within the genetic genealogy community is based on a difference in goals.  People who are looking for the parents of adoptees are looking first and primarily as “sure thing” matches and the bigger the match segment, of course, the better because that means the people are related more closely in time.  For them, smaller segments really are useless.  However, for people who know their recent genealogy and are looking for those brick wall ancestors, several generations back in time, their only hope is utilizing those smaller segments.  This not black and white but shades of grey.  One size does not fit all.  Nor is what we know today the end of the line.  We learn every single day and many of our learning experiences are by working through our own unique genealogical situations – and sharing our discoveries.

On this next spreadsheet, you can see the smaller segments surrounding the larger segments – in other words, in the same match cluster – highlighted in green.  These are the segments that would be discarded as invalid if you were drawing the line at the match threshold.  Some people draw it even higher, at 10 cM.  I’m not being critical of their methodology or saying they are wrong.  It may well work best for them, but discarding small segments is not the only approach and other approaches do work, depending on the goals of the researcher.  I want my 33% IBD segments, thank you very much.

All of the segments highlighted in purple match between at least three cousins.  By checking the other cousins accounts, I can validate that they do all match each other as well, even though I can’t tell this through the Family Tree DNA matrix below the matching threshold.  So, I’ve proven these are valid.  We all received them from our common ancestor.

What about the white rows?  Are those valid matches, from a common ancestor?  We don’t have enough information to make that determination today.

chromosome browser war6

Downloading my data, and confirming segments to this common ancestor allows me to map my own chromosomes.  Now, I know that if someone matches me and any of these three cousins on chromosome 15, for example, between 33,335,760 and 58,455,135 – they are, whether they know it or not, descended from our common ancestral line.

In my opinion, I would think it a shame to discount or throw away all of these matches below 7cM, because you would be discounting 39 of your 52 total matches, or 75% of them.  I would be more conservative in assigning my segments with only one cousin match to any ancestor, but I would certainly note the match and hope that if I added other cousins, that segment would be eventually proven as IBD.

I used positively known cousins in this example because there is no disputing the validity of these matches.  They were known as cousins long before DNA testing.

Breaking Down Brick Walls

This is the same technique utilized to break down brick walls – and the more cousins you have tested, so that you can identify the maximum number of chromosome pieces of a particular ancestor – the better.

I used this same technique to identify Sarah Hickerson in my Thanksgiving Day article, utilizing these same cousins, plus several more.

Hey, just for fun, want to see what chromosome 15 looks like in this much larger sample???

In this case, we were trying to break down a brick wall.  We needed to determine if Sarah Hickerson was the mother of Elijah Vannoy.  All of the individuals in the left “Name” column are proven Vannoy cousins from Elijah, or in one case, William, from another child of Sarah Hickerson.  The individuals in the right “Match” column are everyone in the cousin match group plus the people in green who are Hickerson/Higginson descendants.  William, in green, is proven to descend from Sarah Hickerson and her husband, Daniel Vannoy.

chromosome browser war7

The first part of chromosome 15 doesn’t overlap with the rest.  Buster, David and I share another ancestral line as well, so the match in the non-red section of chromosome 15 may well be from that ancestral line.  It becomes an obvious possibility, because none of the people who share the Vannoy/Hickerson/Higginson DNA are in that small match group.

All of the red colored cells do overlap with at least one other individual in that group and together they form a cluster.  The yellow highlighted cells are the ones over the match threshold.  The 6 Hickerson/Higginson descendants are scattered throughout this match group.

And yes, for those who are going to ask, there are many more Vannoy/Hickerson triangulated groups.  This is just one of over 60 matching groups in total, some with matches well above the match threshold. But back to the chromosome browser wars!

23andMe

This example from 23andMe shows why it’s so very important to verify that your matches also match each other.

chromosome browser war8

Blue and purple match segments are to two of the same cousins that I used in the comparison at Family Tree DNA, who are from my father’s side.  Green is my first cousin from my mother’s side.   Note that on chromosome 11, they both match me on a common segment.  I know by working with them that they don’t match each other on that segment, so while they are both related to me, on chromosome 11, it’s not through the same ancestor.  One is from my father’s side and one is from my mother’s side.  If I hadn’t already known that, determining if they matched each other would be the acid test and would separate them into 2 groups.

23andMe provides you with a tool to see who your matches match that you match too.  That’s a tongue twister.

In essence, you can select any individual, meaning you or anyone that you match, on the left hand side of this tool, and compare them to any 5 other people that you match.  In my case above, I compared myself to my cousins, but if I want to know if my cousin on my mother’s side matches my two cousins on my father’s side, I simply select her name on the left and theirs on the right by using the drop down arrows.

chromosome browser war9

I would show you the results, but it’s in essence a blank chromosome browser screen, because she doesn’t match either of them, anyplace, which tells me, if I didn’t already know, that these two matches are from different sides of my family.

However, in other situations, where I match my cousin Daryl, for example, as well as several other people on the same segment, I want to know how many of these people Daryl matches as well.  I can enter Daryl’s name, with my name and their names in the group of 5, and compare.  23andMe facilitates the viewing or download of the results in a matrix as well, along with the segment data.  You can also download your entire list of matches by requesting aggregated data through the link at the bottom of the screen above or the bottom of the chromosome display.

I find it cumbersome to enter each matches name in the search tool and then enter all of the other matches names as well.  By utilizing the tools at www.dnagedcom.com, you can determine who your matches match as well, in common with you, in one spreadsheet.  Here’s an example.  Daryl in the chart below is my match, and this tool shows you who else she matches that I match as well, and the matching segments.  This allows me to correlate my match with Gwen for example, to Daryl’s match to Gwen to see if they are on the same segments.

chromosome browser war10

As you can see, Daryl and I both match Gwen on a common segment.  On my own chromosome mapping spreadsheet, I match several other people as well at that location, at other vendors, but so far, we haven’t been able to find any common genealogy.

Ancestry.com

At Ancestry.com, I have exactly the opposite problem.  I have lots of people I DNA match, and some with common genealogy, but no tools to prove the DNA match is to the common ancestor.

Hence, this is the crux of the chromosome browser wars.  I’ve just showed you how and why we use chromosome browsers and tools to show if our matches match each other in addition to us and on which segments.  I’ve also illustrated why.  Neither 23andMe nor Family Tree DNA provides perfect tools, which is why we utilize both GedMatch and DNAGedcom, but they do provide tools.  Ancestry provides no tools of this type.

At Ancestry, you have two kinds of genetic matches – ones without tree matches and ones with tree matches.  Pedigree matching is a service that Ancestry provides that the other vendors don’t.  Unfortunately, it also leads people to believe that because they match these people genetically and share a tree, that the tree shown is THE genetic match and it’s to the ancestor shown in the tree.  In fact, if the tree is wrong, either your tree or their tree, and you match them genetically, you will show up as a pedigree match as well.  Even if both pedigrees are right, that still doesn’t mean that your genetic match is through that ancestor.

How many bad trees are at Ancestry percentagewise?  I don’t know, but it’s a constant complaint and there is absolutely nothing Ancestry can do about it.  All they can do is utilize what they have, which is what their customers provide.  And I’m glad they do.  It does make the process of working through your matches much easier. It’s a starting point.  DNA matches with trees that also match your pedigree are shown with Ancestry’s infamous shakey leaf.

In fact, in my Sarah Hickerson article, it was a shakey leaf match that initially clued me that there was something afoot – maybe. I had to shift to another platform (Family Tree DNA) to prove the match however, where I had tools and lots of known cousins.

At Ancestry, I now have about 3000 matches in total, and of those, I have 33 shakey leaves – or people with whom I also share an ancestor in our pedigree charts.  A few of those are the same old known cousins, just as genealogy crazy as me, and they’ve tested at all 3 companies.

The fly in the ointment, right off the bat, is that I noticed in several of these matches that I ALSO share another ancestral line.

Now, the great news is that Ancestry shows you your surnames in common, and you can click on the surname and see the common individuals in both trees.

The bad news is that you have to notice and click to see that information, found in the lower left hand corner of this screen.

chromosome browser war11

In this case, Cook is an entirely different line, not connected to the McKee line shown.

However, in this next case, we have the same individual entered in our software, but differently.  It wasn’t close enough to connect as an ancestor, but close enough to note.  It turns out that Sarah Cook is the mother of Fairwick Claxton, but her middle name was not Helloms, nor was her maiden name, although that is a long-standing misconception that was proven incorrect with her husband’s War of 1812 documents many years ago. Unfortunately, this misinformation is very widespread in trees on the internet.

chromosome browser war12

Out of curiosity, and now I’m sorry I did this because it’s very disheartening – I looked to see what James Lee Claxton/Clarkson’s wife’s name was shown to be on the first page of Ancestry’s advanced search matches.

Despite extensive genealogical and DNA research, we don’t know who James Lee Claxton/Clarkson’s parents are, although we’ve disproven several possibilities, including the most popular candidate pre-DNA testing.  However, James’ wife was positively Sarah Cook, as given by her, along with her father’s name, and by witnesses to their marriage provided when she applied for a War of 1812 pension and bounty land.  I have the papers from the National Archives.

James Lee Claxton’s wife, Sara Cook is identified as follows in the first 50 Ancestry search entries.

Sarah Cook – 4

Incorrect entries:

  • Sarah Cook but with James’ parents listed – 3
  • Sarah Helloms Cook – 2, one with James’ parents
  • Sarah Hillhorns – 15
  • Sarah Cook Hitson – 13, some with various parents for James
  • No wife, but various parents listed for James – 12
  • No wife, no parents – 1

I’d much rather see no wife and no parents than incorrect information.

Judy Russell has expressed her concern about the effects of incorrect trees and DNA as well and we shared this concern with Ancestry during our meeting.

Ancestry themselves in their paper titled “Identifying groups of descendants using pedigrees and genetically inferred relationships in a large database” says, “”As with all analyses relating to DNA Circles™, tree quality is also an important caveat and limitation.”  So Ancestry is aware, but they are trying to leverage and utilize one of their biggest assets, their trees.

This brings us to DNA Circles.  I reviewed Ancestry’s new product release extensively in my Ancestry’s Better Mousetrap article.  To recap briefly, Ancestry gathers your DNA matches together, and then looks for common ancestors in trees that are public using an intelligent ranking algorithm that takes into account:

  1. The confidence that the match is due to recent genealogical history (versus a match due to older genealogical history or a false match entirely).
  2. The confidence that the identified common recent ancestor represents the same person in both online pedigrees.
  3. The confidence that the individuals have a match due to the shared ancestor in question as opposed to from another ancestor or from more distant genealogical history.

The key here is that Ancestry is looking for what they term “recent genealogical history.”  In their paper they define this as 10 generations, but the beta version of DNA Circles only looks back 7 generations today.  This was also reflected in their phasing paper, “Discovering IBD matches across a large, growing database.”

However, the unfortunate effect has been in many cases to eliminate matches, especially from endogamous groups.  By way of example, I lost my Acadian matches in the Ancestry new product release.  They would have been more than 7 generations back, and because they were endogamous, they may have “looked like” IBS segments, if IBS is defined at Ancestry as more than 7 or 10 generations back.  Hopefully Ancestry will tweek this algorithm in future releases.

Ancestry, according to their paper, “Identifying groups of descendants using pedigrees and genetically inferred relationships in a large database,” then clusters these remaining matching individuals together in Circles based on their pedigree charts.  You will match some of these people genetically, and some of them will not match you but will match each other.  Again, according to the paper, “these confidence levels are calculated by the direct-line pedigree size, the number of shared ancestral couples and the generational depth of the shared MRCA couple.”

Ancestry notes that, “using the concordance of two independent pieces of information, meaning pedigree relationships and patterns of match sharing among a set of individuals, DNA Circles can serve as supporting evidence for documented pedigree lines.”  Notice, Ancestry did NOT SAY proof.  Nothing that Ancestry provides in their DNA product constitutes proof.

Ancestry continues by saying that Circles “opens the possibility for people to identify distant relatives with whom they do not share DNA directly but with whom they still have genetic evidence supporting the relationship.”

In other words, Ancestry is being very clear in this paper, which is provided on the DNA Circles page for anyone with Circles, that they are giving you a tool, not “the answer,” but one more piece of information that you can consider as evidence.

joel vannoy circleJoel Vannoy circle2

You can see in my Joel Vannoy circle that I match both of these people both genetically and on their tree.

We, in the genetic genealogy community, need proof.  It certainly could be available, technically – because it is with other vendors and third party sites.

We need to be able to prove that our matches also match each other, and utilizing Ancestry’s tools, we can’t.  We also can’t do this at Ancestry by utilizing third party tools, so we’re in essence, stuck.

We can either choose to believe, without substantiation, that we indeed share a common ancestor because we share DNA segments with them plus a pedigree chart from that common ancestor, or we can initiate a conversation with our match that leads to either or both of the following questions:

  1. Have you or would you upload your raw data to GedMatch?
  2. Have you or would you upload your raw data file to Family Tree DNA?

Let the begging begin!!!

The Problem

In a nutshell, the problem is that even if your Ancestry matches do reply and do upload their file to either Family Tree DNA or GedMatch or both, you are losing most of the potential information available, or that would be available, if Ancestry provided a chromosome browser and matrix type tool.

In other words, you’d have to convince all of your matches and then they would have to convince all of the matches in the circle that they match and you don’t to upload their files.

Given that, of the 44 private tree shakey leaf matches that I sent messages to about 2 weeks ago, asking only for them to tell me the identity of our common pedigree ancestor, so far 2 only of them have replied, the odds of getting an entire group of people to upload files is infinitesimal.  You’d stand a better chance of winning the lottery.

One of the things Ancestry excels at is marketing.

ancestry ad1

If you’ve seen any of their ads, and they are everyplace, they focus on the “feel good” and they are certainly maximizing the warm fuzzy feelings at the holidays and missing those generations that have gone before us.

ancestry ad2

This is by no means a criticism, but it is why so many people do take the Ancestry DNA test. It’s advertised as easy and you’ll learn more about your family.  And you do, no question – you learn about your ethnicity and you get a list of DNA matches, pedigree matches when possible and DNA Circles.

The list of what you don’t get is every bit as important, a chromosome browser and tools to see whether your matches also match each other.  However, most of their customers will never know that.

Judging by the high percentage of inaccurate trees I found at Ancestry in my little experiment relative to the known and documented wife’s name of James Lee Claxton, which was 96%, based on just the first page of 50 search matches, it would appear that about 96% of Ancestry’s clientele are willing to believe something that someone else tells them without verification.  I doubt that it matters whether that information is a tree or a DNA test where they are shown  matches with common pedigree charts and circles.  I don’t mean this to be critical of those people.  We all began as novices and we need new people to become interested in both genealogy and DNA testing.

I suspect that most of Ancestry’s clients, especially new ones, simply don’t have a clue that there is a problem, let alone the magnitude and scope.  How would they?  They are just happy to find information about their ancestor.  And as someone said to me once – “but there are so many of those trees (with a wrong wife’s name), how can they all be wrong?”  Plus, the ads, at least some of them, certainly suggest that the DNA test grows your family tree for you.

ancestry ad3 signoff

The good news in all of this is that Ancestry’s widespread advertising has made DNA testing just part of the normal things that genealogists do.  Their marketing expertise along with recent television programs have served to bring DNA testing into the limelight. The bad news is that if people test at Ancestry instead of at a vendor who provides tools, we, and they, lose the opportunity to utilize those results to their fullest potential.  We, and they, lose any hope of proving an ancestor utilizing DNA.  And let’s face it, DNA testing and genealogy is about collaboration.  Having a DNA test that you don’t compare against others is pointless for genealogy purposes.

When a small group of bloggers and educators visited Ancestry in October, 2014, for what came to be called DNA Day, we discussed the chromosome browser and Ancestry’s plans for their new DNA Circles product, although it had not yet been named at that time.  I wrote about that meeting, including the fact that we discussed the need for a chromosome browser ad nauseum.  Needless to say, there was no agreement between the genetic genealogy community and the Ancestry folks.

When we discussed the situation with Ancestry they talked about privacy and those types of issues, which you can read about in detail in that article, but I suspect, strongly, that the real reason they aren’t keen on developing a chromosome browser lies in different areas.

  1. Ancestry truly believes that people cannot understand and utilize a chromosome browser and the information it provides. They believe that people who do have access to chromosome browsers are interpreting the results incorrectly today.
  2. They do not want to implement a complex feature for a small percentage of their users…the number bantered around informally was 5%…and I don’t know if that was an off-the-cuff number or based on market research. However, if you compare that number with the number of accurate versus inaccurate pedigree charts in my “James Claxton’s wife’s name” experiment, it’s very close…so I would say that the 5% number is probably close to accurate.
  3. They do not want to increase their support burden trying to explain the results of a chromosome browser to the other 95%. Keep in mind the number of users you’re discussing. They said in their paper they had 500,000 DNA participants. I think it’s well over 700,000 today, and they clearly expect to hit 1 million in 2015. So if you utilize a range – 5% of their users are 25,000-50,000 and 95% of their users are 475,000-950,000.
  4. Their clients have already paid their money for the test, as it is, and there is no financial incentive for Ancestry to invest in an add-on tool from which they generate no incremental revenue and do generate increased development and support costs. The only benefit to them is that we shut up!

So, the bottom line is that most of Ancestry’s clients don’t know or care about a chromosome browser.  There are, however, a very noisy group of us who do.

Many of Ancestry’s clients who purchase the DNA test do so as an impulse purchase with very little, if any, understanding of what they are purchasing, what it can or will do for them, at Ancestry or anyplace else, for that matter.

Any serious genealogist who researched the autosomal testing products would not make Ancestry their only purchase, especially if they could only purchase one test.  Many, if not most, serious genealogists have tested at all three companies in order to fish in different ponds and maximize their reach.  I suspect that most of Ancestry’s customers are looking for simple and immediate answers, not tools and additional work.

The flip side of that, however, if that we are very aware of what we, the genetic genealogy industry needs, and why, and how frustratingly lacking Ancestry’s product is.

Company Focus

It’s easy for us as extremely passionate and focused consumers to forget that all three companies are for-profit corporations.  Let’s take a brief look at their corporate focus, history and goals, because that tells a very big portion of the story.  Every company is responsible first and foremost to their shareholders and owners to be profitable, as profitable as possible which means striking the perfect balance of investment and expenditure with frugality.  In corporate America, everything has to be justified by ROI, or return on investment.

Family Tree DNA

Family Tree DNA was the first one of the companies to offer DNA testing and was formed in 1999 by Bennett Greenspan and Max Blankfeld, both still principles who run Family Tree DNA, now part of Gene by Gene, on a daily basis.  Family Tree DNA’s focus is only on genetic genealogy and they have a wide variety of products that produce a spectrum of information including various Y DNA tests, mitochondrial, autosomal, and genetic traits.  They are now the only commercial company to offer the Y STR and mitochondrial DNA tests, both very important tools for genetic genealogists, with a great deal of information to offer about our ancestors.

In April 2005, National Geographic’s Genographic project was announced in partnership with Family Tree DNA and IBM.  The Genographic project, was scheduled to last for 5 years, but is now in its 9th year.  Family Tree DNA and National Geographic announced Geno 2.0 in July of 2012 with a newly designed chip that would test more than 12,000 locations on the Y chromosome, in addition to providing other information to participants.

The Genographic project provided a huge boost to genetic genealogy because it provided assurance of legitimacy and brought DNA testing into the living room of every family who subscribed to National Geographic magazine.  Family Tree DNA’s partnership with National Geographic led to the tipping point where consumer DNA testing became mainstream.

In 2011 the founders expanded the company to include clinical genetics and a research arm by forming Gene by Gene.  This allowed them, among other things, to bring their testing in house by expanding their laboratory facilities.  They have continued to increase their product offerings to include sophisticated high end tests like the Big Y, introduced in 2013.

23andMe

23andMe is also privately held and began offering testing for medical and health information in November 2007, initially offering “estimates of predisposition for more than 90 traits ranging from baldness to blindness.”  Their corporate focus has always been in the medical field, with aggregated customer data being studied by 23andMe and other researchers for various purposes.

In 2009, 23andMe began to offer the autosomal test for genealogists, the first company to provide this service.  Even though, by today’s standards, it was very expensive, genetic genealogists flocked to take this test.

In 2013, after several years of back and forth with 23andMe ultimately failing to reply to the FDA, the FDA forced 23andMe to stop providing the medical results.  Clients purchasing the 23andMe autosomal product since November of 2013 receive only ethnicity results and the genealogical matching services.

In 2014, 23andMe has been plagued by public relations issues and has not upgraded significantly nor provided additional tools for the genetic genealogy community, although they recently formed a liaison with My Heritage.

23andMe is clearly focused on genetics, but not primarily genetic genealogy, and their corporate focus during this last year in particular has been, I suspect, on how to survive, given the FDA action.  If they steer clear of that landmine, I expect that we may see great things in the realm of personalized medicine from them in the future.

Genetic genealogy remains a way for them to attract people to increase their data base size for research purposes.  Right now, until they can again begin providing health information, genetic genealogists are the only people purchasing the test, although 23andMe may have other revenue sources from the research end of the business

Ancestry.com

Ancestry.com is a privately held company.  They were founded in the 1990s and have been through several ownership and organizational iterations, which you can read about in the wiki article about Ancestry.

During the last several years, Ancestry has purchased several other genealogy companies and is now the largest for-profit genealogy company in the world.  That’s either wonderful or terrible, depending on your experiences and perspective.

Ancestry has had an on-again-off-again relationship with DNA testing since 2002, with more than one foray into DNA testing and subsequent withdrawal from DNA testing.  If you are interested in the specifics, you can read about them in this article.

Ancestry’s goal, as it is with all companies, is profitability.  However, they have given themselves a very large black eye in the genetic genealogy community by doing things that we consider to be civically irresponsible, like destroying the Y and mitochondrial DNA data bases.  This still makes no sense, because while Ancestry spends money on one hand to acquire data bases and digitize existing records, on the other hand, they wiped out a data base containing tens of thousands of irreplaceable DNA records, which are genealogy records of a different type.  This was discussed at DNA Day and the genetic genealogy community retains hope that Ancestry is reconsidering their decision.

Ancestry has been plagued by a history of missteps and mediocrity in their DNA products, beginning with their Y and mitochondrial DNA products and continuing with their autosomal product.  Their first autosomal release included ethnicity results that gave many people very high percentages of Scandinavian heritage.  Ancestry never acknowledged a problem and defended their product to the end…until the day when they announced an update titled….a whole new you.  They are marketing geniuses.  While many people found their updated product much more realistic, not everyone was happy.  Judy Russell wrote a great summary of the situation.

It’s difficult, once a company has lost their credibility, for them to regain it.

I think Ancestry does a bang up job of what their primary corporate goal is….genealogy records and subscriptions for people to access those records. I’m a daily user.  Today, with their acquisitions, it would be very difficult to be a serious genealogist without an Ancestry subscription….which is of course what their corporate goal has been.

Ancestry does an outstanding job of making everything look and appear easy.  Their customer interface is intuitive and straightforward, for the most part. In fact, maybe they have made both genealogy and genetic genealogy look a little too easy.  I say this tongue in cheek, full well knowing that the ease of use is how they attract so many people, and those are the same people who ultimately purchase the DNA tests – but the expectation of swabbing and the answer appearing is becoming a problem.  I’m glad that Ancestry has brought DNA testing to so many people but this success makes tools like the chromosome browser/matrix that much more important – because there is so much genealogy information there just waiting to be revealed.  I also feel that their level of success and visibility also visits upon them the responsibility for transparency and accuracy in setting expectations properly – from the beginning – with the ads. DNA testing does not “grow your tree” while you’re away.

I’m guessing Ancestry entered the DNA market again because they saw a way to sell an additional product, autosomal DNA testing, that would tie people’s trees together and provide customers with an additional tool, at an additional price, and give them yet another reason to remain subscribed every year.  Nothing wrong with that either.  For the owners, a very reasonable tactic to harness a captive data base whose ear you already have.

But Ancestry’s focus or priority is not now, and never has been, quality, nor genetic genealogy.  Autosomal DNA testing is a tool for their clients, a revenue generation source for them, and that’s it.  Again, not a criticism.  Just the way it is.

In Summary

As I look at the corporate focus of the three players in this space, I see three companies who are indeed following their corporate focus and vision.  That’s not a bad thing, unless the genetic genealogy community focus finds itself in conflict with the results of their corporate focus.

It’s no wonder that Family Tree DNA sponsors events like the International DNA Conference and works hand in hand with genealogists and project administrators.  Their focus is and always has been genetic genealogy.

People do become very frustrated with Family Tree DNA from time to time, but just try to voice those frustrations to upper management at either 23andMe or Ancestry and see how far you get.  My last helpdesk query to 23andMe submitted on October 24th has yet to receive any reply.  At Family Tree DNA, I e-mailed the project administrator liaison today, the Saturday after Thanksgiving, hoping for a response on Monday – but I received one just a couple hours later – on a holiday weekend.

In terms of the chromosome browser war – and that war is between the genetic genealogy community and Ancestry.com, I completely understand both positions.

The genetic genealogy community has been persistent, noisy, and united.  Petitions have been created and signed and sent to Ancestry upper management.  To my knowledge, confirmation of any communications surrounding this topic with the exception of Ancestry reaching out to the blogging and education community, has never been received.

This lack of acknowledgement and/or action on the issues at hand frustrates the community terribly and causes reams of rather pointed and very direct replies to Anna Swayne and other Ancestry employees who are charged with interfacing with the public.  I actually feel sorry for Anna.  She is a very nice person.  If I were in her position, I’d certainly be looking for another job and letting someone else take the brunt of the dissatisfaction.  You can read her articles here.

I also understand why Ancestry is doing what they are doing – meaning their decision to not create a chromosome browser/match matrix tool.  It makes sense if you sit in their seat and now have to look at dealing with almost a million people who will wonder why they have to use a chromosome browser and or other tools when they expected their tree to grow while they were away.

I don’t like Ancestry’s position, even though I understand it, and I hope that we, as a community, can help justify the investment to Ancestry in some manner, because I fully believe that’s the only way we’ll ever get a chromosome browser/match matrix type tool.  There has to be a financial benefit to Ancestry to invest the dollars and time into that development, as opposed to something else.  It’s not like Ancestry has additional DNA products to sell to these people.  The consumers have already spent their money on the only DNA product Ancestry offers, so there is no incentive there.

As long as Ancestry’s typical customer doesn’t know or care, I doubt that development of a chromosome browser will happen unless we, as a community, can, respectfully, be loud enough, long enough, like an irritating burr in their underwear that just won’t go away.

burr

The Future

What we “know” and can do today with our genomes far surpasses what we could do or even dreamed we could do 10 years ago or even 5 or 2 years ago.  We learn everyday.

Yes, there are a few warts and issues to iron out.  I always hesitate to use words like “can’t,” “never” and “always” or to use other very strongly opinionated or inflexible words, because those words may well need to be eaten shortly.

There is so much more yet to be done, discovered and learned.  We need to keep open minds and be willing to “unlearn” what we think we knew when new and better information comes along.  That’s how scientific discovery works.  We are on the frontier, the leading edge and yes, sometimes the bleeding edge.  But what a wonderful place to be, to be able to contribute to discovery on a new frontier, our own genes and the keys to our ancestors held in our DNA.

Anzick (12,707-12,556), Ancient One, 52 Ancestors #42

anzick burial location

His name is Anzick, named for the family land, above, where his remains were found, and he is 12,500 years old, or more precisely, born between 12,707 and 12,556 years before the present.  Unfortunately, my genealogy software is not prepared for a birth year with that many digits.  That’s because, until just recently, we had no way to know that we were related to anyone of that age….but now….everything has changed ….thanks to DNA.

Actually, Anzick himself is not my direct ancestor.  We know that definitively, because Anzick was a child when he died, in present day Montana.

anzick on us map

Anzick was loved and cherished, because he was smeared with red ochre before he was buried in a cave, where he would be found more than 12,000 years later, in 1968, just beneath a layer of approximately 100 Clovis stone tools, shown below.  I’m sure his parents then, just as parents today, stood and cried as the laid their son to rest….never suspecting just how important their son would be some 12,500 years later.

anzick clovis tools

From 1968 until 2013, the Anzick family looked after Anzick’s bones, and in 2013, Anzick’s DNA was analyzed.

DNA analysis of Anzick provided us with his mitochondrial haplogroup,  D4h3a, a known Native American grouping, and his Y haplogroup was Q-L54, another known Native American haplogroup.  Haplogroup Q-L54 itself is estimated to be about 16,900 years old, so this finding is certainly within the expected range.  I’m not related to Anzick through Y or mitochondrial DNA.

Utilizing the admixture tools at GedMatch, we can see that Anzick shows most closely with Native American and Arctic with a bit of east Siberian.  This all makes sense.

Anzick MDLP K23b

Full genome sequencing was performed on Anzick, and from that data, it was discovered that Anzick was related to Native Americans, closely related to Mexican, Central and South Americans, and not closely related to Europeans or Africans.  This was an important discovery, because it in essence disproves the Solutrean hypothesis that Clovis predecessors emigrated from Southwest Europe during the last glacial maximum, about 20,000 years ago.

anzick matches

The distribution of these matches was a bit surprising, in that I would have expected the closest matches to be from North America, in particular, near to where Anzick was found, but his closest matches are south of the US border.  Although, in all fairness, few people in Native tribes in the US have DNA tested and many are admixed.

This match distribution tells us a lot about population migration and distribution of the Native people after they left Asia, crossed Beringia on the land bridge, now submerged, into present day Alaska.

This map of Beriginia, from the 2008 paper by Tamm et all, shows the migration of Native people into (and back from) the new world.

beringia map

Anzick’s ancestors crossed Beringia during this time, and over the next several thousand years, found their way to Montana.  Some of Anzick’s relatives found their way to Mexico, Central and South America.  The two groups may have split when Anzick’s family group headed east instead of south, possibly following the edges of glaciers, while the south-moving group followed the coastline.

Recently, from Anzick’s full genome data, another citizen scientist extracted the DNA locations that the testing companies use for autosomal DNA results, created an Anzick file, and uploaded the file to the public autosomal matching site, GedMatch.  This allowed everyone to see if they matched Anzick.  We expected no, or few, matches, because after all, Anzick was more than 12,000 years old and all of his DNA would have washed out long ago due to the 50% replacement in every generation….right?  Wrong!!!

What a surprise to discover fairly large segments of DNA matching Anzick in living people, and we’ve spent the past couple of weeks analyzing and discussing just how this has happened and why.  In spite of some technical glitches in terms of just how much individual people carry of the same DNA Anzick carried, one thing is for sure, the GedMatch matches confirm, in spades, the findings of the scientists who wrote the recent paper that describes the Anzick burial and excavation, the subsequent DNA processing and results.

For people who carry known Native heritage, matches, especially relatively large matches to Anzick, confirm not only their Native heritage, but his too.

For people who suspect Native heritage, but can’t yet prove it, an Anzick match provides what amounts to a clue – and it may be a very important clue.

In my case, I have proven Native heritage through the Micmac who intermarried with the Acadians in the 1600s in Nova Scotia.  Given that Anzick’s people were clearly on a west to east movement, from Beringia to wherever they eventually wound up, one might wonder if the Micmac were descended from or otherwise related to Anzick’s people.  Clearly, based on the genetic affinity map, the answer is yes, but not as closely related to Anzick as Mexican, Central and South Americans.

After several attempts utilizing various files, thresholds and factors that produced varying levels of matching to Anzick, one thing is clear – there is a match on several chromosomes.  Someplace, sometime in the past, Anzick and I shared a common ancestor – and it was likely on this continent, or Beringia, since the current school of thought is that all Native people entered the New World through this avenue.  The school of thought is not united in an opinion about whether there was a single migration event, or multiple migrations to the new word.  Regardless, the people came from the same base population in far northeast Asia and intermingled after arriving here if they were in the same location with other immigrants.

In other words, there probably wasn’t much DNA to pass around.  In addition, it’s unlikely that the founding population was a large group – probably just a few people – so in very short order their DNA would be all the same, being passed around and around until they met a new population, which wouldn’t happen until the Europeans arrived on the east side of the continent in the 1400s.  The tribes least admixed today are found south of the US border, not in the US.  So it makes sense that today the least admixed people would match Anzick the most closely – because they carry the most common DNA, which is still the same DNA that was being passed around and around back then.

Many of us with Native ancestors do carry bits and pieces of the same DNA as Anzick.  Anzick can’t be our ancestor, but he is certainly our cousin, about 500 generations ago, using a 25 year generation, so roughly our 500th cousin.  I had to laugh at someone this week, an adoptee who said, “Great, I can’t find my parents but now I have a 12,500 year old cousin.”  Yep, you do!  The ironies of life, and of genealogy, never fail to amaze me.

Utilizing the most conservative matching routine possible, on a phased kit, meaning one that combines the DNA shared by my mother and myself, and only that DNA, we show the following segment matches with Anzick.

Chr Start Location End Location Centimorgans (cM) SNPs
2 218855489 220351363 2.4 253
4 1957991 3571907 2.5 209
17 53111755 56643678 3.4 293
19 46226843 48568731 2.2 250
21 35367409 36761280 3.7 215

Being less conservative produces many more matches, some of which are questionable as to whether they are simply convergence, so I haven’t utilized the less restrictive match thresholds.

Of those matches above, the one on chromosomes 17 matches to a known Micmac segment from my Acadian lines and the match on chromosome 2 also matches an Acadian line, but I share so many common ancestors with this person that I can’t tell which family line the DNA comes from.

There are also Anzick autosomal matches on my father’s side.  My Native ancestry on his side reaches back to colonial America, in either Virginia or North Carolina, or both, and is unproven as to the precise ancestor and/or tribe, so I can’t correlate the Anzick DNA with proven Native DNA on that side.  Neither can I associate it with a particular family, as most of the Anzick matches aren’t to areas on my chromosome that I’ve mapped positively to a specific ancestor.

Running a special utility at GedMatch that compared Anzick’s X chromosome to mine, I find that we share a startlingly large X segment.  Sometimes, the X chromosome is passed for generations intact.

Interestingly enough, the segment 100,479,869-103,154,989 matches a segment from my mother exactly, but the large 6cM segment does not match my mother, so I’ve inherited that piece of my X from my father’s line.

Chr Start Location End Location Centimorgans (cM) SNPs
X 100479869 103154989 1.4 114
X 109322285 113215103 6.0 123

This tells me immediately that this segment comes from one of the pink or blue lines on the fan chart below that my father inherited from his mother, Ollie Bolton, since men don’t inherit an X chromosome from their father.  Utilizing the X pedigree chart reduces the possible lines of inheritance quite a bit, and is very suggestive of some of those unknown wives.

olliex

It’s rather amazing, if you think about it, that anyone today matches Anzick, or that we can map any of our ancestral DNA that both we and Anzick carry to a specific ancestor.

Indeed, we do live in exciting times.

Honoring Anzick

On a rainy Saturday in June, 2014, on a sagebrush hillside in Montana, in Native parlance, our “grandfather,” Anzick was reburied, bringing his journey full circle.  Sarah Anzick, a molecular biologist, the daughter of the family that owns the land where the bones were found, and who did part of the genetic discovery work on Anzick, returns the box with his bones for reburial.

anzick bones

More than 50 people, including scientists, members of the Anzick family and representatives of six Native American tribes, gathered for the nearly two-hour reburial ceremony. Tribe members said prayers, sang songs, played drums and rang bells to honor the ancient child. The bones were placed in the grave and sprinkled with red ocher, just like when his parents buried him some 12,500 years before.

Participants at the reburial ceremony filled in the grave with handfuls, then shovelfuls of dirt and covered it with stones. A stick tied with feathers marks Anzick’s final resting place.

Sarah Anzick tells us that, “At that point, it stopped raining. The clouds opened up and the sun came out. It was an amazing day.”

I wish I could have been there.  I would have, had I known.  After all, he is part of me, and I of him.

anzick grave'

Welcome to the family, Anzick, and thank you, thank you oh so much, for your priceless, unparalleled gift!!!

tobacco

If you want to read about the Anzick matching journey of DNA discovery, here are the articles I’ve written in the past two weeks.  It has been quite a roller coaster ride, but I’m honored and privileged to be doing this research.  And it’s all thanks to an ancient child named Anzick.

Utilizing Ancient DNA at Gedmatch

Analyzing the Native American Anzick Clovis Native American Results

New Native American Mitochondrial DNA Haplogroups Extrapolated from Anzick Match Results

Ancient DNA Matching, A Cautionary Tale

More Ancient DNA Samples for Comparison

More Ancient DNA Samples For Comparison

Felix Chandrakumar has prepared and added three additional ancient DNA kits to GedMatch.  Thanks Felix!  This is a wonderful service you’re performing for the genetic genealogy community!

  • The Linearbandkeramik (LBK) sample, also referenced as “Stuttgart,” reflecting where it was discovered in Germany.  This individual was an early farmer dating from about 7,500 years ago and was one of the samples analyzed for the paper, Ancient genomes suggest three ancestral populations for present-day Europeans. Kit F999916
  • The La Brana-Arintero sample from Leon, Spain, about 7000 years old, represents a pre-agricultural European human genome – in other words, before the agriculturists from the Near East arrived. In an article at Science Daily, they have reconstructed his face. Original academic article available here. Kit F999915
  • The Mal’ta sample, from Siberia, about 24,000 years of age. The results were discussed in article, Native American Gene Flow – Europe?, Asia and the Americas, and the original article is available here. Kit F999914

These kits, along with the ones listed earlier, give us the opportunity to compare our own DNA with that of ancient people in specific populations.  It’s like taking a step back in time and seeing if we carry any of the same small segments as these people did – suggesting of course that we descend from the same population.

This Ancient European DNA map by Richard Stevens shows the European locations where ancient DNA has been retrieved.

ancient dna map2

Recent discussion has focused on determining what matches to these specimens actually mean to genetic genealogists today.  We obviously don’t have that answer at this point.  We know that, due to their age, these samples are not close relatives in terms of genealogy generations, but in some cases, we find that we have matches far larger than one would expect to be found utilizing the 50% washout per generation math.

Endogamy, especially in a closed population such as Native Americans is certainly one explanation.  That doesn’t explain the European matches however – either to Anzick, the Native American specimen, nor Europeans to the European samples.  The higher no-call rate in the autosomal files can contribute as well, but wouldn’t account for all matches.  In some cases, maybe everyone carries the same DNA because the population carries that DNA in very high rates – but the population carries the DNA in very high rates because the ancient ancestors did as well…so this is a bit of circular logic.  All that said, we’re still left wondering what is real and what is Memorex, so to speak?

Ancient DNA is changing our understanding of the human past, and that of our ancestors.  It allows us a connection to the ancient people that is tangible, parts of them found in us today, as unbelievable as it seems.

When Svante Paabo discovered that modern Europeans all carry pieces of Neanderthal DNA, he too was struck by what I’ll call “the disbelief factor,” thinking, of course, that it can’t possibly be true.  He discussed this at length in his book, Neanderthal Man, In Search of Lost Genomes, and the steps taken by his team to prove that the matches weren’t in error or due to some problem with the ancient genome reconstruction process.  Indeed, all Europeans and Asians carry both Neanderthal and Denisovan DNA, and by the same process of the DNA being carried by the entire population at one point, which must be the avenue for contemporary humans to carry other ancient DNA as well.  As we find individual matches to small pieces of DNA with these matches, how much of that is “real” versus convergence or a result of no-calls in the ancient files?

In that vein, I find this article from Dienekes Anthropology Blog quite interesting,  found in the ASHG Titles of Interest from the upcoming Conference in October in San Diego, CA.

Reducing pervasive false positive identical-by-descent segments detected by large-scale pedigree analysis. E. Y. Durand, N. Eriksson, C. Y. McLean.

“Analysis of genomic segments shared identical-by-descent (IBD) between individuals is fundamental to many genetic applications, from demographic inference to estimating the heritability of diseases. A large number of methods to detect IBD segments have been developed recently. However, IBD detection accuracy in non-simulated data is largely unknown. In principle, it can be evaluated using known pedigrees, as IBD segments are by definition inherited without recombination down a family tree. We extracted 25,432 genotyped European individuals containing 2,952 father-mother-child trios from the 23andMe, Inc. dataset. We then used GERMLINE, a widely used IBD detection method, to detect IBD segments within this cohort. Exploiting known familial relationships, we identified a false positive rate over 67% for 2-4 centiMorgan (cM) segments, in sharp contrast with accuracies reported in simulated data at these sizes. We show that nearly all false positives arise due to allowing switch errors between haplotypes when detecting IBD, a necessity for retrieving long (> 6 cM) segments in the presence of imperfect phasing. We introduce HaploScore, a novel, computationally efficient metric that enables detection and filtering of false positive IBD segments on population-scale datasets. HaploScore scores IBD segments proportional to the number of switch errors they contain. Thus, it enables filtering of spurious segments reported due to GERMLINE being overly permissive to imperfect phasing. We replicate the false IBD findings and demonstrate the generalizability of HaploScore to alternative genotyping arrays using an independent cohort of 555 European individuals from the 1000 Genomes project. HaploScore can be readily adapted to improve the accuracy of segments reported by any IBD detection method, provided that estimates of the genotyping error rate and switch error rate are available.”

I’m pleased to see that they are addressing smaller segments, in the 2cM-4cM range, because those are the ranges some are finding in matches to these ancient genomes.  A few matches are even larger.

Of course, all of this ancient matching has caused an upsurge in interest in the cultures and populations of these ancient people whose DNA we carry.

I find this graphic very interesting from the paper, Toward a new history and geography of human genes informed by ancient DNA, just published this month, by Joseph Pickrell and David Reich.  This map, which shows the population movement into and out of geographic regions of the world in the past, is especially interesting in that several back migrations are shown into Africa.  I’ve never seen the “history of the world in population migration” summed up quite so succinctly before, but it helps us understand why certain DNA is found in specific locations.

population man

Copyright @2014 Elsevier Ltd, Trends in Genetics, 2014, 30, 377-389DOI: (10/1016/j.tig.2014.07.007

As we find and fully sequence additional ancient DNA specimens, we’ll be able to better understand how the ancient populations were related to each other, and then, how we descend from each of them.

This is a fascinating age of personal discovery!

Ancient DNA Matching – A Cautionary Tale

egg

I hope that all of my readers realize that you are literally watching science hatch.  We are on the leading, and sometimes bleeding edge, of this new science of genetic genealogy.  Because many of these things have never been done before, we have to learn by doing and experimenting.  Because I blog about this, these experiments are “in public,” so there is no option of a private “oops.”  Fortunately, I’m not sensitive about these kinds of things.  Plus, I think people really enjoy coming along for the ride of discovery.  I mean, where else can you do that?  It’s really difficult to get a ride-along on the space shuttle!

One of the best pieces of advice I ever got was from someone who was taken from my life far too early.  I had made a mistake of some sort…don’t even remember what…and he gave me a card that said, “The only people who don’t make mistakes are the people who don’t try.”

This isn’t an “oops” moment.  More like an “aha” moment.  Or more precisely, a “huh” moment.  It falls in the “Houston, we’ve got a problem” category.

So, this week’s new discovery is that there seems to be some inconsistency in the matching to the Anzick kit at GedMatch.  Before I go any further, I want to say very clearly that this is in no way a criticism of anyone or any tool.  Every person involved is a volunteer and we would not be making any of these steps forward, including a few backwards, without these wonderful volunteers and tools.

I have reached out to the people involved and asked for their help to unravel this mystery, and I’m sharing the story with you, partly so you can understand what is involved, and the process, partly so that you don’t inadvertently encounter the same kinds of issues and draw unrealistic or incorrect conclusions, and partly so you can help.  If there has been any common theme in all of my articles in the past week or so about the ancient DNA articles, it has been that we really don’t understand what conclusions to draw yet…and we still don’t.  So don’t.

Let’s introduce the players here.

The Players

Felix Chandrakumar has very graciously prepared the various ancient DNA files and uploaded them to GedMatch.  Felix has written a number of DNA analysis tools as well.

John Olson is one of the two volunteers who created and does everything at Gedmatch, plus works a full time job.  By the way, in case you’re not aware, this is a contribution site, meaning they depend on your financial contributions to function, purchase hardware, servers, etc.  If you use this site, periodically scroll down and click on the donate button.  We, as a community, would be lost without John and his partner.

David Pike is a long time genetic genealogist who I have had the pleasure of working with on a number of Native American and related topics over the years. He also has created several genetic genealogy tools to deal with autosomal DNA. David prepared the Anzick files for some private work we were doing several months ago, so he has experience with this DNA as well.  Dr. Pike has a great deal of experience analyzing the endogamous population of Newfoundland, which is also admixed with Native Americans.

Marie Rundquist, also a long time genetic genealogist who specializes in both technology and Acadian history along with genetic genealogy.  Acadians are proven to be admixed with Native Americans.  Marie shares my deep interest and commitment to Native American study and genetics.  Furthermore, Marie and I also share ancestors and co-administer several related projects.   As you might imagine, Marie and I took this opportunity immediately to see if she and her mother share any of Anzick’s segments with me and my mother.

So, a big thank you to all of these people.

The Mystery

When Felix originally e-mailed me about the Anzick kit being uploaded to GedMatch, as you might imagine, I stopped doing whatever I was doing and immediately went to study Anzick and the other ancient DNA kits.

I wrote about this experience in the article, “Utilizing Ancient DNA at GedMatch.”

As part of that process, I not only ran Anzick’s kit utilizing the “one to many” option, I also compared my own kit to Anzick’s.  My proven Native lines descend through my mother, so I ran her kit against Anzick’s as well, at the same thresholds, and I combined the two results to see where mother and I overlapped.

I showed these overlaps in the article, along with which genealogy lines they matched by utilizing my ancestor matching spreadsheet.

Everything was hunky dory…for then.

Day 2

The next day, I received a note from Felix that the Anzick kit may not have been fully tokenized at GedMatch previously, so I reran the Anzick “one to all” comparison and wrote about those results in the second article, “Analyzing the Native American Clovis Ancient Results.”  Because it wasn’t yet fully processed originally, the second results produced more matches, not fewer.

I wasn’t worried about the one to one comparison of Anzick to my own kit, because one to one comparisons are available immediately, while one to many comparisons are not, per the GedMatch instructions.

“Once you have loaded your data, you will be able to use some features of the site within a minute or so. Additional batch processing, which usually takes a couple of days, must complete before you can use some of the tools comparing you to everyone in the data pool.”

So, everything was stlll hunky dory.

Day 3

The next day, Marie and I had a few minutes, sometime between 2 and 3AM, and no, I’m not kidding.  We decided to compare results.  I decided it would be quicker to run the match again at GedMatch than to sort through my Master spreadsheet, into which I had copied the results and added other information.  So, I did a second download of the Anzick comparison, utilizing the exact same thresholds (200 SNPs, 2cM, and the rest left at the default,) and added them to a spreadsheet that Marie and I were passing back and forth, and sent them to Marie.  I noticed that there seemed to be fewer matches, but by then it was after 3AM and I decided to follow up on that later.

Not so hunky dory…but I didn’t know it yet.

Day 4

The following day, Dr. Ann Turner (MD), also a long-time genetic genealogist, posted the following comment on the article.

“These results, finding “what appear to be contemporary matches for the Anzick child”, seemed very counter-intuitive to me, so I asked John Olson of GEDMatch to look under the hood a bit more. It turns out the ancient DNA sequence has many no-calls, which are treated as universal matches for segment analysis. Another factor which should be examined is whether some of the matching alleles are simply the variants with the highest frequency in all populations. If so, that would also lead to spurious matching segments. It may not be appropriate to apply tools developed for genetic genealogy to ancient DNA sequences like this without a more thorough examination of the underlying data.”

I had been aware of the no-calls due to the work that Dr. David Pike did back in March with the Anzick raw data files, but according to David, that shouldn’t affect the results.

Here’s what Dr. Pike, a Professor of Mathematics, had to say:

“Yes, these forensic samples have very high No-Call rates, which may give rise to more false matches than we would normally experience.  Also, be aware that false matches are more prone to occur when using reduced thresholds (such as 100 SNPs and 1 cM) and unphased data.  In this case I don’t think there’s any way around using low thresholds, simply because we’re looking for very small blocks of DNA (probably nobody alive today will have any large matching blocks with the Anzick child).

On the assumption that there will be a nearly constant noise ratio, meaning that most people will have about the same number of false matches with the Anzick child, those who are from the same gene pool should have an increased number of real matches.  So by comparing the total amount of matching DNA, it ought to be possible to gauge people’s affinity with Anzick’s gene pool.”

Here are Felix’s comments about no-calls as well:

“Personally, no calls are fine as long as there are more SNPs matching above the threshold level because the possibility of errors occurring exactly on no-call positions for all the matches in all their matching segments is impossible.”

Courtesy of Felix, we’ll see an example of how no calls intersperse in  a few minutes.

If no-calls were causing spurious matches in the Anzick kit, you’d expect to see the same for the other ancient DNA kits.  I know that the Denisovan and Neanderthal kits also have many no-calls, and based on the nature of ancient DNA, I’m sure all of them do.  So, if no calls are the culprit, they should be affecting matches to the other kits in the same way, and they aren’t.

Hunky-doryness is being replaced by a nonspecific nagging feeling…same one I used to get when my teenagers were up to something.

Day 5

A day or so later, Felix uploaded file F999913 to replace F999912 with the complete SNPs from all of the companies.  The original 999912 kit only included the SNP locations utilized by Family Tree DNA.  Felix added the SNPs utilized by 23and Me not utilized at Family Tree DNA, and the ones from Ancestry as well.  This is great news for anyone who tested at those two companies, but I had utilized my kit from Family Tree DNA, so for me, there should be no difference at all.

I later asked Felix if he had changed anything else in the file, and he said that he had not.  He provided extensive documentation about what he had done.

I waited until kit F999912 was deleted to be sure tokenizing was complete for F999913 and re-compared the data again.  As expected, Anzick’s one to all had more matches than before, because additional people were included due to the added SNPs from 23andMe and Ancestry.

Some of Anzick’s matches are in the contemporary range, at 3.1 estimated generations, with the largest cM segment of 22.8 and total cMs of 202.8.

anzick 999913

These relatively large matches cause Felix to question whether the sample is actually ancient, based on these relatively large segments.  I addressed my feelings on this in the article, Ancient DNA Matches – What Do They Mean?

Marie and Dr. Pike, both with extensive experience with admixed populations addressed this as well.  Marie commented,

“Native DNA found in the Anzick sample hasn’t changed all of that much and may still be found in modern, Native American populations, and that if people have Native American ancestry, they’ll match to it.”

Dr. Pike says:

“I agree with Marie on this… within endogamous populations, there is an increased likelihood of blocks of DNA being preserved over lengthy time frames.  Moreover, even if a block of DNA gets cut up via recombination, within an endogamous population the odds of some parts of the block later reuniting in a person’s DNA are higher than otherwise.  And it exaggerates the closeness of [the] relationship that gets predicted when comparing people.

I have seen something similar within the Newfoundland & Labrador Family Finder Project, whereby lots of people are sharing small blocks of DNA, likely as a result of DNA from the early colonists still circulating among the modern gene pool.

As an anecdotal example, I have a semi-distant relative (with ancestry from Newfoundland) at 23andMe who shares 3 blocks of DNA with my father, 2 with my mother and 5 five me.  As you can imagine, the relative is predicted to be a closer cousin to me than she is to either of my parents!

It doesn’t take an endogamous or isolated population to see this effect.

It can also happen in families involving cousin marriages too, although that would be more pronounced and not quite the same thing as we’re discussing with respect to ancient DNA.”

This addition of other companies SNPs should not affect my matches with Anzick because my kits are both from FTDNA and won’t utilize the added SNPs.

However, I ran my and my mother’s matches again, and we had a significantly different outcome than either of the previous times.

I utilized the same threshold for all downloads and those are the only values I changed – 200 SNPs and 2cM, leaving the other values at default, for all Anzick comparisons to my mother and my kits.

I am not hunky-dory anymore.

The Heartburn

These matches, which should be the same in all three downloads, produced significantly different results.

Here are the number of matches at the same threshold comparing me and Mom to the Anzick file:

Me and Anzick

  • original download 999912 – 47 matches
  • second download 999912 – 21 matches
  • 999913 – 35 matches

Mom and Anzick

  • original download 999912 – 63
  • second download 999912 – 37
  • 999913 – 36

And no, the 36 /35 that mom and I have for 999913 are not all the same.

Kit Number Matches Between Me, Mother and Anzick
#1-F999912 original download 19
#2-F999912 second download 6
#3-F999913 11

Of those various downloads, the following grid shows which ones matched each other.

#1 to #2 #2 to #3 #1 to #3 All 3
# of Matches 6 2 3 2

So, comparing the first download to the last download, of the 19 original matches, we lost 16 matches.  In the third download, we gained 8 matches and only 3 remained as common matches. So of 30 total matches between my mother, myself and Anzick, in two downloads that should have been exactly the same, only 3 matches held, or 10%.

Obviously, something is wrong, but what, and where?  At that point, I asked Marie to download her and her mother’s results again too, and she experienced the same issue.

Clearly a problem exists someplace.  That’s the question I asked Felix, John and David to help answer.

I realize that this spreadsheet it very long, and I apologize, but I think this issue is much easier to see visually.  I’ve compiled the matches by color and shade to make looking at them relatively easy.

My matches to the Anzick kit are in shades of pink – the first match download being the lightest and the last one to kit F999913 being the darkest.  Mother is green, same shading scheme.

The three columns to the right show the matching segments for each download – shaded in green.  You can easily see which ones line up, meaning which ones match consistently across all three downloads.  There aren’t many.  They should all match.

anzick me mom problem

Obviously this led to many questions that I asked of the various players involved.

My first thought was that perhaps a matching algorithm change occurred in GedMatch, but John assured me that he had made no changes.

Next question was whether or not Felix changed something other than adding the 23andMe and Ancestry SNPs.  He had not.

Felix was kind enough to explain about bunching and to do some analysis on the files.

“When you have low thresholds, make sure you don’t allow errors. For example, at 200 SNPs, the default ‘Mismatch Evaluation window’ and in GEDMatch is same as SNP threshold and ‘Mismatch-Bunching limit’ is half of mismatch evaluation window. So, at 200 cM, you are allowing 1 error every 100 SNPs apart from no-calls.

I did some analysis on your phased mother’s kit, PF6656M1 so that at least we know that it is an IBD for one generation.  The spreadsheet (below) are segments I found at 2 cM/200 SNPs threshold without allowing any errors.”

Kit PF6656M1 is one single kit created by phasing my data against my mother’s so that we don’t have to run both kits.  I had not utilized the phased kit previously, so I was interested in his results.

felix anzick

The results above confirm chromosome matches, 2, 17, 19 and 21, but introduce a new match on chromosome 4.  This match was present in the original download, but not in the second or third download, so once again, we have disparate data, except the thresholds Felix used were at a different level.

One of the more interesting things that Felix included is the no-call match information, the three columns to the right.  I want to show what the no-calls look like.  There are not huge segments that are blank and are being called as matches because they are no-calls, when they shouldn’t be.  No calls are scattered like salt and pepper.  In fact, no calls happen in every kit and they are called as matches so they don’t in fact disrupt a valid match string, potentially making it too small to be considered a match.  Of course, ancient DNA has more no-calls that contemporary DNA kits.

Below are the first few match positions from chromosome 2 where mother, Anzick and I have a confirmed match across all downloads.  The genotype shows you that both kits match.

felix no calls

For consistency, I ran the same kits that Felix ran, PF6656M1 and F999913, with the original thresholds I had used, and found the following:

Chr Start Location End Location Centimorgans (cM) SNPs
1 31358221 33567640 2.0 261
2 218855489 220351363 2.4 253
4 1957991 3571907 2.5 209
5 2340730 2982499 2.3 200
17 53111755 56643678 3.4 293
19 46226843 48568731 2.2 250
21 35367409 36761280 3.7 215

This introduces chromosomes 1 and 5, not shown above.   The chromosome 1 match was shown in the first and second download, but not the third, and the chromosome 5 match was shown in the first download only, but not the second or third.

Can you see me beating my head against the wall yet??

In a fit of apparent insanity, I decided to try, once again, an individual download of Anzick compared to my mother and to me, but not utilizing the phased kit – the original F6656 and F9141, and at the original thresholds, for consistency.  I wanted to see if the matches were the same now as they were a day or so ago.  They should be exact.  This first one is mine.

me second 999913

What you should see are two identical downloads.  I have color coded the rows so you can see easily – and what you should see are candy-cane stripes – one red and one white for every match location.

That’s not what we’re seeing.  The kits are the same, the match parameters are the same, but the results are not.  Once again, the downloads don’t match.

I did another match on mother and Anzick, and her results were consistent between the first and second match to kit F999913.

mom second 999913

The begs the next question.  Have mother’s results always been consistent, suggesting a problem with my kit?

I sorted all of her downloads, and no, they are not consistent, except for the first and second download matches to kit F999913, shown above.  The inconsistencies show up in both mother and my kits, although not in the same locations.  Recall also that Marie had the same issue.

In Summary

Something is wrong, someplace.  I know that sounds intuitively obvious – NOW.  But it wasn’t initially and I wouldn’t even have suspected a problem without running the second and third downloads, quite unintentionally.  Most people never do that, because once you’ve done the match, you have no reason to ever match to that particular person again.  Given that, you’ll never know if a problem exists.

So, the only Anzick GedMatch matches I have any confidence in at all, at this point, are the few that are consistent between all of the downloads, and I didn’t add the fourth download into the mix.  I don’t’ see any point because I’ve pretty much concluded that until we determine where the issue resides, that I won’t have confidence in the results.

The next question that comes to mind, and that I can’t answer, is whether or not this issue is present in contemporary matching kits – or if this is somehow an ancient DNA problem – although I don’t know quite how that could be – since matching is matching.

I haven’t saved any matches that I’ve run to other people in spreadsheets, so I can’t go back and see if a GedMatch match today produces the exact same results as a previous match.

Clearly there is no diagnosis or solution in this summary.  We are not yet hunky dory.

What You Can Do

  1. Run your Anzick and ancient DNA matches multiple times, at the same exact thresholds, on different days, to see if your results are consistent or inconsistent. Same kit, same thresholds, the results should be identical.
  2. If you have some saved GedMatch matches with contemporary people, and you are positive of the match thresholds used, please run them again to see if the results are identical. They should be.
  3. No drawing of or jumping to conclusions, please, especially about ancient DNA:) It’s a journey and we are fellow pilgrims!

If your results are not consistent, please document the problem and let the appropriate person know.  I don’t want to overwhelm John at GedMatch but I’m concerned at this point that the problem may not be isolated to ancient DNA matching since the issue seems to extend to Marie’s results as well.

If your results, especially to Anzick, from previous matches to now are consistent, that’s worth knowing too.  Please add a comment to that effect.

Thoughts and ideas are welcome.

Ancient DNA Matches – What Do They Mean?

The good news is that my three articles about the Anzick and other ancient DNA of the past few days have generated a lot of interest.

The bad news is that it has generated hundreds of e-mails every day – and I can’t possibly answer them all personally.  So, if you’ve written me and I don’t reply, I apologize and  I hope you’ll understand.  Many of the questions I’ve received are similar in nature and I’m going to answer them in this article.  In essence, people who have matches want to know what they mean.

Q – I had a match at GedMatch to <fill in the blank ancient DNA sample name> and I want to know if this is valid.

A – Generally, when someone asks if an autosomal match is “valid,” what they really mean is whether or not this is a genealogically relevant match or if it’s what is typically referred to as IBS, or identical by state.  Genealogically relevant samples are referred to as IBD, or identical by descent.  I wrote about that in this article with a full explanation and examples, but let me do a brief recap here.

In genealogy terms, IBD is typically used to mean matches over a particular threshold that can be or are GENEALOGICALLY RELEVANT.  Those last two words are the clue here.  In other words, we can match them with an ancestor with some genealogy work and triangulation.  If the segment is large, and by that I mean significantly over the threshold of 700 SNPs and 7cM, even if we can’t identify the common ancestor with another person, the segment is presumed to be IBD simply because of the math involved with the breakdown of segment into pieces.  In other words, a large segment match generally means a relatively recent ancestor and a smaller segment means a more distant ancestor.  You can readily see this breakdown on this ISOGG page detailing autosomal DNA transmission and breakdown.

Unfortunately, often smaller segments, or ones determined to be IBS are considered to be useless, but they aren’t, as I’ve demonstrated several times when utilizing them for matching to distant ancestors.  That aside, there are two kinds of IBS segments.

One kind of IBS segment is where you do indeed share a common ancestor, but the segment is small and you can’t necessarily connect it to the ancestor.  These are known as population matches and are interpreted to mean your common ancestor comes from a common population with the other person, back in time, but you can’t find the common ancestor.  By population, we could mean something like Amish, Jewish or Native American, or a country like Germany or the Netherlands.

In the cases where I’ve utilized segments significantly under 7cM to triangulate ancestors, those segments would have been considered IBS until I mapped them to an ancestor, and then they suddenly fell into the IBD category.

As you can see, the definitions are a bit fluid and are really defined by the genealogy involved.

The second kind of IBS is where you really DON’T share an ancestor, but your DNA and your matches DNA has managed to mutate to a common state by convergence, or, where your Mom’s and Dad’s DNA combined form a pseudo match, where you match someone on a segment run long enough to be considered a match at a low level.  I discussed how this works, with examples, in this article.  Look at example four, “a false match.”

So, in a nutshell, if you know who your common ancestor is on a segment match with someone, you are IBD, identical by descent.  If you don’t know who your common ancestor is, and the segment is below the normal threshold, then you are generally considered to be IBS – although that may or may not always be true.  There is no way to know if you are truly IBS by population or IBS by convergence, with the possible exception of phased data.

Data phasing is when you can compare your autosomal DNA with one or both parents to determine which half you obtained from whom.  If you are a match by convergence where your DNA run matches that of someone else because the combination of your parents DNA happens to match their segment, phasing will show that clearly.  Here’s an example for only one location utilizing only my mother’s data phased with mine.  My father is deceased and we have to infer his results based on my mother’s and my own.  In other words, mine minus the part I inherited from my mother = my father’s DNA.

My Result My Result Mother’s Result Mother’s Result Father’s Inferred Result Father’s Inferred Result
T A T G A

In this example of just one location, you can see that I carry a T and an A in that location.  My mother carries a T and a G, so I obviously inherited the T from her because I don’t have a G.  Therefore, my father had to have carried at least an A, but we can’t discern his second value.

This example utilized only one location.  Your autosomal data file will hold between 500,000 and 700,000 location, depending on the vendor you tested with and the version level.

You can phase your DNA with that of your parent(s) at GedMatch.  However, if both of your parents are living, an easier test would be to see if either of your parents match the individual in question.  If neither of your parents match them, then your match is a result of convergence or a data read error.

So, this long conversation about IBD and IBS is to reach this conclusion.

All of the ancient specimens are just that, ancient, so by definition, you cannot find a genealogy match to them, so they are not IBD.  Best case, they are IBS by population.  Worse case, IBS by convergence.  You may or may not be able to tell the difference.  The reason, in my example earlier this week, that I utilized my mother’s DNA and only looked at locations where we both matched the ancient specimens was because I knew those matches were not by convergence – they were in fact IBS by population because my mother and I both matched Anzick.

ancient compare5

Q – What does this ancient match mean to me?

A – Doggone if I know.  No, I’m serious.  Let’s look at a couple possibilities, but they all have to do with the research you have, or have not, done.

If you’ve done what I’ve done, and you’ve mapped your DNA segments to specific ancestors, then you can compare your ancient matching segments to your ancestral spreadsheet map, especially if you can tell unquestionably which side the ancestral DNA matches.  In my case, shown above, the Clovis Anzik matched my mother and me on the same segment and we both matched Cousin Herbie.  We know unquestionably who our common ancestor is with cousin Herbie – so we know, in our family line, which line this segment of DNA shared with Anzick descends through.

ancient compare6

If you’re not doing ancestor mapping, then I guess the Anzick match would come in the category of, “well, isn’t that interesting.”  For some, this is a spiritual connection to the past, a genetic epiphany.  For other, it’s “so what.”

Maybe this is a good reason to start ancestor mapping!  This article tells you how to get started.

Q – Does my match to Anzick mean he is my ancestor?

A – No, it means that you and Anzick share common ancestry someplace back in time, perhaps tens of thousands of years ago.

Q – I match the Anzick sample.  Does this prove that I have Native American heritage? 

A – No, and it depends.  Don’t you just hate answers like this?

No, this match alone does not prove Native American heritage, especially not at IBS levels.  In fact, many people who don’t have Native heritage match small segments?  How can this be?  Well, refer to the IBS by convergence discussion above.  In addition, Anzick child came from an Asian population when his ancestors migrated, crossing from Asia via Beringia.  That Eurasian population also settled part of Europe – so you could be matching on very small segments from a common population in Eurasia long ago.  In a paper just last year, this was discussed when Siberian ancient DNA was shown to be related to both Native Americans and Europeans.

In some cases, a match to Anzick on a segment already attributed to a Native line can confirm or help to confirm that attribution.  In my case, I found the Anzick match on segments in the Lore family who descend from the Acadians who were admixed with the Micmac.  I have several Anzick match segments that fit that criteria.

A match to Anzick alone doesn’t prove anything, except that you match Anzick, which in and of itself is pretty cool.

Q – I’m European with no ancestors from America, and I match Anzick too.  How can that be?

A – That’s really quite amazing isn’t it.  Just this week in Nature, a new article was published discussing the three “tribes” that settled or founded the European populations.  This, combined with the Siberian ancient DNA results that connect the dots between an ancient population that contributed to both Europeans and Native Americans explains a lot.

3 European Tribes

If you think about it, this isn’t a lot different than the discovery that all Europeans carry some small amount of Neanderthal and Denisovan DNA.

Well, guess what….so does Anzick.

Here are his matches to the Altai Neanderthal.

Chr Start Location End Location Centimorgans (cM) SNPs
2 241484216 242399416 1.1 138
3 19333171 21041833 2.6 132
6 31655771 32889754 1.1 133

He does not match the Caucasus Neanderthal.  He does, however, match the Denisovan individual on one location.

Chr Start Location End Location Centimorgans (cM) SNPs
3 19333171 20792925 2.1 107

Q – Maybe the scientists are just wrong and the burial is not 12,500 years old,  maybe just 100 years old and that’s why the results are matching contemporary people.

A – I’m not an archaeologist, nor do I play one…but I have been closely involved with numerous archaeological excavations over the past decade with The Lost Colony Research Group, several of which recovered human remains.  The photo below is me with Anne Poole, my co-director, sifting at one of the digs.

anne and me on dig

There are very specific protocols that are followed during and following excavation and an error of this magnitude would be almost impossible to fathom.  It would require  kindergarten level incompetence on the part of not one, but all professionals involved.

In the Montana Anzick case, in the paper itself, the findings and protocols are both discussed.  First, the burial was discovered directly beneath the Clovis layer where more than 100 tools were found, and the Clovis layer was undisturbed, meaning that this is not a contemporary burial that was buried through the Clovis layer.  Second, the DNA fragmentation that occurs as DNA degrades correlated closely to what would be expected in that type of environment at the expected age based on the Clovis layer.  Third, the bones themselves were directly dated using XAD-collagen to 12,707-12,556 calendar years ago.  Lastly, if the remains were younger, the skeletal remains would match most closely with Native Americans of that region, and that isn’t the case.  This graphic from the paper shows that the closest matches are to South Americans, not North Americans.

anzick matches

This match pattern is also confirmed independently by the recent closest GedMatch matches to South Americans.

Q – How can this match from so long ago possibly be real?

A – That’s a great question and one that was terribly perplexing to Dr. Svante Paabo, the man who is responsible for producing the full genome sequence of the first, and now several more, Neanderthals.  The expectation was, understanding autosomal DNA gets watered down by 50% in every generation though recombination, that ancient genomes would be long gone and not present in modern populations.  Imagine Svante’s surprise when he discovered that not only isn’t true, but those ancient DNA segmetns are present in all Europeans and many Asians as well.  He too agonized over the question about how this is possible, which he discussed in this great video.  In fact he repeated these tests over and over in different ways because he was convinced that modern individuals could not carry Neanderthal DNA – but all those repeated tests did was to prove him right.  (Paabo’s book, Neanderthal Man, In Search of Lost Genomes is an incredible read that I would highly recommend.)

What this means is that the population at one time, and probably at several different times, had to be very small.  In fact, it’s very likely that many times different pockets of the human race was in great jeopardy of dying out.  We know about the ones that survived.  Probably many did perish leaving no descendants today.  For example, no Neanderthal mitochondrial DNA has been found in any living or recent human.

In a small population, let’s say 5 males and 5 females who some how got separated from their family group and founded a new group, by necessity.  In fact, this could well be a description of how the Native Americans crossed Beringia.  Those 5 males and 5 females are the founding population of the new group.  If they survive, all of the males will carry the men’s haplogroups – let’s say they are Q and C, and all of the descendants will carry the mitochondrial haplogroups of the females – let’s say A, B, C, D and X.

There is a very limited amount of autosomal DNA to pass around.  If all of those 10 people are entirely unrelated, which is virtually impossible, there will be only 10 possible combinations of DNA to be selected from.  Within a few generations, everyone will carry part of those 10 ancestor’s DNA.  We all have 8 ancestors at the great-grandparent level.  By the time those original settlers’ descendants had great-great-grandparents – of which each one had 16, at least 6 of those original people would be repeated twice in their tree.

There was only so much DNA to be passed around.  In time, some of the segments would no longer be able to be recombined because when you look at phasing, the parents DNA was exactly the same, example below.  This is what happens in endogamous populations.

My Result My Result Mother’s Result Mother’s Result Father’s Result Father’s  Result
T T T T T T

Let’s say this group’s descendants lived without contact with other groups, for maybe 15,000 years in their new country.  That same DNA is still being passed around and around because there was no source for new DNA.  Mutations did occur from time to time, and those were also passed on, of course, but that was the only source of changed DNA – until they had contact with a new population.

When they had contact with a new population and admixture occurred, the normal 50% recombination/washout in every generation began – but for the previous 15,000 years, there had been no 50% shift because the DNA of the population was, in essence, all the same.  A study about the Ashkenazi Jews that suggests they had only a founding population of about 350 people 700 years ago was released this week – explaining why Ashkenazi Jewish descendants have thousands of autosomal matches and match almost everyone else who is Ashkenazi.  I hope that eventually scientists will do this same kind of study with Anzick and Native Americans.

If the “new population” we’ve been discussing was Native Americans, their males 15,000 year later would still carry haplogroups Q and C and the mitochondrial DNA would still be A, B, C, D and X.  Those haplogroups, and subgroups formed from mutations that occurred in their descendants, would come to define their population group.

In some cases, today, Anzick matches people who have virtually no non-Native admixture at the same level as if they were just a few generations removed, shown on the chart below.

anzick gedmatch one to all

Since, in essence, these people still haven’t admixed with a new population group, those same ancient DNA segments are being passed around intact, which tells us how incredibly inbred this original small population must have been.  This is known as a genetic bottleneck.

The admixture report below is for the first individual on the Anzick one to all Gedmatch compare at 700 SNPs and 7cM, above.  In essence, this currently living non-admixed individual still hasn’t met that new population group.

anzick1

If this “new population” group was Neanderthal, perhaps they lived in small groups for tens of thousands of years, until they met people exiting Africa, or Denisovans, and admixed with them.

There weren’t a lot of people anyplace on the globe, so by virtue of necessity, everyone lived in small population groups.  Looking at the odds of survival, it’s amazing that any of us are here today.

But, we are, and we carry the remains, the remnants of those precious ancestors, the Denisovans, the Neanderthals and Anzick.  Through their DNA, and ours, we reach back tens of thousands of years on the human migration path.  Their journey is also our journey.  It’s absolutely amazing and it’s no wonder people have so many questions and such a sense of enchantment.  But it’s true – and only you can determine exactly what this means to you.

Utilizing Ancient DNA at GedMatch

Mummy of 6 month old boy found in Greenland

It has been a wonderful week for those of us following ancient DNA full genome sequencing, because now we can compare our own results to those of the ancient people found whose DNA has been fully sequenced, including one Native American.

Felix Chandrakumar has uploaded the autosomal files of five ancient DNA specimens that have been fully sequenced to GedMatch.  Thanks Felix.

When news of these sequences first hit the academic presses, I was wishing for a way to compare our genomes – and now my wish has come true.

Utilizing GedMatch’s compare one to all function, I ran all of the sequences individually and found, surprisingly, that there are, in some cases, matches to contemporary people today.  I dropped the cM measure to 1 for both autosomal and X.

Please note that because these are ancient DNA sequences, they will all have some segments missing and none can be expected to be entirely complete.  Still, these sequences are far better than nothing.

1.  Montana Anzick at GedMatch

This is the only clearly Native American sample.

http://www.y-str.org/2014/09/clovis-anzick-dna.html

F999912

9-27-2014 – Please note that kit F999912 has been replaced by kit F999913.

10-23-2014 – Please note that kit F999913 has been replaced by kit F999919.

No matches at 1cM in the compare to all.  This must be because the SNP count is still at default thresholds, in light of information discovered later in this article.

Update – as it turns out, this kit was not finished processing when I did the one to one compare.  After it finished, the results were vastly different.  See this article for results.

2.  Paleo Eskimo from Greenland at GedMatch

http://www.y-str.org/2013/12/palaeo-eskimo-2000-bc-dna.html

F999906

Thirty-nine matches with segments as large at 3.8.  One group of matches appears to be a family.  One of these matches is my cousin’s wife.  That should lead to some interesting conversation around the table this holiday season!  All of these matches, except 1, are on the X chromosome.  This must be a function of these segments being passed intact for many generations.

I wrote about some unusual properties of X chromosomal inheritance and this seems to confirm that tendency in the X chromosome, or the matching thresholds are different at GedMatch for the X.

3.  Altai Neanderthal at GedMatch

http://www.y-str.org/2013/08/neanderthal-dna.html

F999902

One match to what is obviously another Neaderthal entry.

4.  Russian Causasus Neanderthal at GedMatch

Another contribution from the Neanderthal Genome Project.

http://www.y-str.org/2014/09/mezmaiskaya-neanderthal-dna.html

F999909

No matches.

5.  Denisova at GedMatch

http://www.y-str.org/2013/08/denisova-dna.html

F999903

Two matches, one to yet another ancient entry and one to a contemporary individual on the X chromosome.

But now, for the fun part.

My Comparison

Before I start this section, I want to take a moment to remind everyone just how old these ancient segments are.

  • Anzick – about 12,500 years old
  • Paleo-Eskimo – about 4,000 years old
  • Altai Neanderthal – about 50,000 years old
  • Russian Caucasus Neanderthal – about 29,000 years old
  • Denisova – about 30,000 years old

In essence, the only way for these segments to survive intact to today would have been for them to enter the population of certain groups, as a whole, to be present in all of the members of that group, so that segment would no longer be divided and would be passed intact for many generation, until that group interbred with another group who did not carry that segment.  This is exactly what we see in endogamous populations today, such as the Askenazi Jewish population who is believed, based on their common shared DNA, to have descended from about 350 ancestors about 700 years ago.  Their descendants today number in the millions.

So, let’s see what we find.

I compared by own kit at GedMatch utilizing the one to one comparison feature, beginning with 500 SNPs and 1cM, dropping the SNP values to 400, then 300, then 200, until I obtained a match of some sort, if I obtained a match at all.

Typically in genetic genealogy, we’re looking for genealogy matches, so the default matching thresholds are set relatively high.  In this case, I’m looking for deep ancestral connections, if they exist, so I was intentionally forcing the thresholds low.  I’m particularly interested in the Anzick comparison, in light of my Native American and First Nations heritage.

The definition of IBS, identical by state, vs IBD, identical by descent segments varies by who is talking and in what context, but in essence, IBD means that there is a genealogy connection in the past several generations.

IBS means that the genealogy connection cannot be found and the IBS match can be a function of coming from a common population at some time in the past, or it can be a match by convergence, meaning that your DNA just happened to mutate to the same state as someone else’s.  If this is the case, then you wouldn’t expect to see multiple segments matching the same person and you would expect the matching segments to be quite short.  The chances of hundreds of SNPs just happening to align becomes increasingly unlikely the longer the matching SNP run.

So, having said that, here are my match results.

Anzick

I had 2 matches at 400 SNPs, several at 300 and an entire list at 200, shown below.

Chr Start Location End Location Centimorgans (cM) SNPs
1 6769350 7734985 1.7 232
1 26552555 29390880 1.9 264
1 31145273 33730360 2.7 300
1 55655110 57069976 1.9 204
1 71908934 76517614 2.8 265
1 164064635 165878596 2.8 264
1 167817718 171330902 3.3 466
1 186083870 192208998 4.2 250
2 98606363 100815734 1.4 256
2 171132725 173388331 2.0 229
2 218855489 220373983 2.5 261
3 128892631 131141396 1.7 263
3 141794591 143848459 2.5 207
4 1767539 3571907 2.7 235
4 70345811 73405268 2.5 223
5 2340730 2982499 2.3 200
5 55899022 57881001 2.3 231
5 132734528 134538202 1.9 275
5 137986213 140659207 1.7 241
6 34390761 36370969 1.8 293
8 17594903 18464321 1.9 200
8 23758017 25732105 1.7 240
8 109589884 115297391 1.9 203
9 122177526 124032492 1.6 229
10 101195132 102661955 1.2 264
10 103040561 105596277 1.3 304
10 106135611 108371247 1.5 226
12 38689229 41184500 1.6 247
13 58543514 60988948 1.6 220
13 94528801 95252127 1.0 277
14 60929984 62997711 1.8 255
14 63724184 65357663 1.7 201
14 72345879 74206753 1.7 263
15 36850933 38329491 2.7 238
16 1631282 2985328 2.5 273
16 11917282 13220406 3.7 276
16 15619825 17324720 3.1 305
16 29085336 31390250 1.3 263
16 51215026 52902771 3.4 224
17 52582669 56643678 4.7 438
19 11527683 13235913 1.7 203
19 15613137 16316773 1.2 204
19 46195917 49338412 3.3 397
20 17126434 18288231 2.1 225
21 35367409 36969215 4.1 254
21 42399499 42951171 1.6 233
22 33988022 35626259 5.0 289

In my case, I’m particularly fortunate, because my mother tested her DNA as well.  By process of elimination, I can figure out which of my matches are through her, and then by inference, which are through my father or are truly IBS by convergence.

I carry Native heritage on both sides, but my mother’s is proven to specific Native ancestors where my father’s is only proven to certain lines and not yet confirmed through genealogy records to specific ancestors.

Because I had so many matches, quite to my surprise, I also compared my mother’s DNA to the Anzick sample, combined the two results and put them in a common spreadsheet, shown below.  White are my matches.  Pink are Mom’s matches, and the green markers are on the segments where we both match the Anzick sample, confirming that my match is indeed through mother.

ancient compare

We’ll work with this information more in a few minutes.

Paleo

At 200 SNP level, 2 segments.

1 26535949 27884441 1.1 258
2 127654021 128768822 1.2 228

My mother matches on 9 segments, but neither of the two above, so they are either from my father’s side or truly IBS by convergence.

Altai Neanderthal

ancient compare2

Russian Neanderthal

Neither my mother nor I have any matches at 100SNPs and 1cM.

Denisovan

I have one match.

Chr Start Location End Location Centimorgans (cM) SNPs
4 8782230 9610959 1.2 100

My mother matches 2 segments at 100 SNPs but neither match is the same as my segment.

Matching to Ancestral Lines

I’ve been mapping my DNA to specific ancestors utilizing the genealogy information of matches and triangulation for some time.  This consists of finding common ancestors with your matches.  Finding one person who matches you and maps to a common ancestor on a particular segment consists of a hint.  Finding two that share the same ancestral line and match you and each other on the same segment is confirmation – hence, the three of you triangulate.  More than three is extra gravy:)

I have also recorded other relevant information in my matches file, like the GedMatch Native chromosomal comparisons when I wrote “The Autosomal Me” series about hunting for my Native chromosomal segments.

So, after looking at the information above, it occurred to me that I should add this ancestral match information to my matches spreadsheet, just for fun, if nothing else.

I added these matches, noted the source as GedMatch and then sorted the results, anxious to see what we might find.  Would at least one of these segments fall into the proven Native segments or the matches to people who also descend from those lines?

What I found was both astonishing and confusing….and true to form to genealogy, introduced new questions.

I have extracted relevant matching groups from my spreadsheet and will discuss them and why they are relevant.  You can click on any of the images to see a larger image.

ancient compare3

This first set of matches is intensely interesting, and equally as confusing.

First, these matches are to both me and mother, so they are confirmed through my mother’s lines.  In case anyone notices, yes, I did switch my mother’s line color to white and mine to pink to be consistent with my master match spreadsheet coloration.

Second, both mother and I match the Anzick line on the matches I’ve utilized as examples.

Third, both 23andMe and Dr. Doug McDonald confirmed the segments in red as Native which includes the entire Anzick segment.

Fourth, utilizing the Gedmatch admixture tools, mother and I had this range in common.  I described this technique in “The Autosomal Me” series.

Fifth, these segments show up for two distinct genealogy lines that do not intersect until my grandparents, the Johann Michael Miller line AND the Acadian Lore line.

Sixth, the Acadian Lore line is the line with proven Native ancestors.

Seventh, the Miller line has no Native ancestors and only one opportunity for a Native ancestor, which is the unknown wife of Philip Jacob Miller who married about 1750 to a women rumored to be Magdalena Rochette, but research shows absolutely no source for that information, nor any Rochette family anyplace in any proximity in the same or surrounding counties to the Miller family.  The Miller’s were Brethren.  Furthermore, there is no oral history of a Native ancestor in this line, but there have been other hints along the way, such as the matching segments of some of the “cousins” who show as Native as well.

Eighth, this makes my head hurt, because this looks, for all the world, like Philip Jacob Miller who was living in Bedford County, PA when he married about 1750 may have married someone related to the Acadian lines who had intermarried with the Micmac.  While this is certainly possible, it’s not a possibility I would ever have suspected.

Let’s see what else the matches show.

ancient compare4

In this matching segment Mom and I both match Emma, who descends from Marie, a MicMac woman.  Mom’s Anzik match is part of this same segment.

ancient compare5

In this matching segment, Mom and I both match cousin Denny who descends from the Lore line who is Acadian and confirmed to have MicMac ancestry.  Mom’s Anzik segments all fit in this range as well.

ancient compare6

In this matching segment, cousin Herbie’s match to Mom and I falls inside the Anzick segments of both Mom and I.

ancient compare7

More matching to the proven Miller line.

ancient compare8

This last grouping with Mom is equally as confusing at the first.  Mom and I both match cousin Denny on the Lore side, proven Acadian.

Mom and I both match the Miller side too, and the Anzik for both of us falls dead center in these matches.

There are more, several more matches, that also indicate these same families, but I’m not including them because they don’t add anything not shown in these examples.  Interestingly enough, there are no pointers to other families, so this isn’t something random.  Furthermore, on my father’s side, as frustrating as it is, here are no Anzick matches that correlate with proven family lines.  ARGGHHHHHH……

On matches that I don’t share with mother, there is one of particular interest.

ancient compare9

You’ll notice that the Anzik and the Paleo-Greenland samples match each other, as well as me.  This is my match, and by inference, not through mother.  Unfortunately, the other people in this match group don’t know their ancestors or we can’t identify a common ancestor.

Given the genetic genealogy gold standard of checking to see if your autosomal matches match each other, I went back to GedMatch to see if the Paleo-Greenland kit matched the Clovis Anzik kit on this segment, and indeed, they do, plus many more segments as well.  So, at some time, in some place, the ancestors of these two people separated by thousands of miles were related to each other.  Their common ancestor would have either been in Asia or in the Northern part of Canada if the Paleo people from Greenland entered from that direction.

Regardless, it’s interesting, very interesting.

What Have I Learned?

Always do experiments.  You never know what you’ll find.

I’m much more closely related to the Anzick individual than I am to the others. This isn’t surprising given my Native heritage along with the endogamous culture of the Acadians.

My relationship level to these ancient people is as follows:

Lived Years Ago Relatedness Comments
Montana Anzick 12,500 107.4cM at 200 SNP level Confirmed to Lore (Acadian) and Miller, but not other lines
Greenland Paleo 4,000 2.3cM at 200 SNP level No family line matches, does match to Anzick in one location
Altai Neanderthal 50,000 2.1cM at 200 SNP level No family line matches
Russian Neanderthal 29,000 0
Denisovan 30,000 1.2cM at 200 SNP No family line matches

The Lores and the Millers

Looking further at the Lore and Miller lines, there are only two options for how these matching segments could have occurred.  There are too many for them all to be convergence, so we’ll have to assume that they are indeed because we shared a common population at some time and place.

The nature of how small the segments are testify that this is not a relatively recent common ancestor, but how “unrecent” is open to debate.  Given that Neanderthal and Denisovan ancient segments are found in all Europeans today, it’s certainly possible for these segments to be passed intact, even after thousands of years.

The confirmations to the Lore line come through proven Lore cousins and also through other proven Acadian non-specific matches.  This means that the Acadian population is highly endogamous and when I find an Acadian match, it often means that I’m related through many ancestors many times.  This, of course, increases the opportunity for the DNA to be passed forward, and decreases the opportunity for it to be lost in transmission, but it also complicates the genealogy greatly and makes determining which ancestor the DNA segment came from almost impossible.

However, I think we are safe to say the segments are from the Acadian population, although my assumption would be that they are from the Native Ancestors and not the French, given the high number of Anzick matches, Anzick being proven to be Native.  Having said that, that assumption may not be entirely correct.

The Miller line is relatively well documented and entirely from Germany/Switzerland, immigrating in the early 1700s, with the exception of the one unknown wife in the first generation married in the US.  Further examination would have to be done to discover if any of the matches came through Johann Michael Miller’s sons other than Philip Jacob Miller, my ancestor.  There are only three confirmed children, all sons.  If this segment shows up in Johann Michael Miller’s line not associated with son Philip Jacob Miller, then we would confirm that indeed the segment came from Europe and not a previously unknown Native or mixed wife of Philip Jacob.

Bottom Line

So, what’s the bottom line here?  I know far more than I did.  The information confirms, yet again, the Acadian Native lines, but it introduces difficult questions about the Miller line.  I have even more tantalizing questions for which I have no answers today, but I tell you what, I wouldn’t trade this journey along the genetic pathway with all of its unexpected bumps, rocks, slippery slopes and crevices for anything!!  That’s why it’s called an adventure!

Finding Native American Ethnic Results in Germanic People

I’m often asked about the significance of small percentages of autosomal DNA in results.  Specifically, the small percentages are often of Native American or results that would suggest Native admixture.  One of the first questions I always ask is whether or not the individual has Germanic or eastern European admixture.

Why?

Take a look at this map of the Invasion of the Roman Empire.  See the Huns and their path?

Hun map

It’s no wonder we’re so admixed.

Here’s a map of the Hunnic empire at its peak under Attila between the years 420-469.

Hun emplire

But that wasn’t the end of the Asian invasions.  The Magyars, who settled in Hungary arrived from Asia as well, in the 800s and 900s, as shown on this map from LaSalle University.

magyar map

Since both the Hungarians and some Germanic people descend from Asian populations, as do Native Americans, albeit thousands of years apart, it’s not unrealistic to expect that, as populations, they share a genetic connection.

Therefore, when people who carry heritage from this region of the world show small amounts of Native or Asian origin, I’m not surprised.  However, for Americans, trying to sort out their Native ethnic heritage, this is most unhelpful.

Let’s take a look at the perfect example candidate.  This man is exactly half Hungarian and half German.  Let’s see what his DNA results say, relative to any Asian or Native heritage, utilizing the testing companies and the free admixture tools at www.gedmatch.com.

He has not tested at Ancestry, but at Family Tree DNA, his myOrigins report 96% European, 4% Middle Eastern.  At 23andMe in speculative view, he shows 99.7 European and .2 sub-saharan African.

Moving to the admixture tools at GedMatch, MDLP is not recommended for Asian or Native ancestry, so I have excluded that tool.

Eurogenes K13 is the most recently updated admixture tool, so let’s take a look at that one first.

Eurogenes K13

 JK Eurogenes K13 v2

Eurogenes K13 showed 7% West Asian, which makes perfect sense considering his heritage, but it might be counted as “Native” in other circumstances, although I would certainly be very skeptical about counting it as such.

However, East Asian, Siberian and Amerindian would all be amalgamated into the Native American category, for a combined percentage of 1.31.

jk eurogenes k13 chart

However, selecting the “admixture proportions by chromosome” view shows something a bit different.  The cumulative percentages, by chromosome equate to 10.10%.  Some researchers mistakenly add this amount and use that as their percentage of Native ancestry.  This is not the case, because those are the portions of 100% of each individual chromosome, and the total would need to be divided by 22 to obtain the average value across all chromosomes.  The total is irrelevant, and the average may not reflect how the developer determines the amount of admixture because chromosomes are not the same size nor carry the same number of SNPs.  Questions relative to the functional underpinnings of each tool should be addressed to the developers.

Dodecad

I understand that there is a newer version of Dodecad, but that it has not been submitted to GedMatch for inclusion, per a discussion with GedMatch.  I can’t tell which of the Dodecad versions on GedMatch is the most current, so I ran the results utilizing both v3 and 12b.

jk dodecad v3

jk dodecad v3 chart

I hope v3 is not the most current, because it does not include any Native American category or pseudocategory – although there is a smattering of Northeast Asian at .27% and Southwest Asian at 1%.

Dodecad 12b below

jk dodecad 12b

The 12b version does show .52% Siberian and 2.6% Southwest Asian, although I’m not at all sure the Southwest Asian should be included.

HarappaWorld

jk harappaworld

jk harappaworld chart

Harappaworld shows .09 Siberian, .27% American (Native American), .23% Beringian and 1.8% Southwest Asian, although I would not include Southwest Asian in the Native calculation.

In Summary

Neither Family Tree DNA nor 23andMe find Native ancestry in our German/Hungarian tester, but all 3 of the admixture tools at Gedmatch find either small amounts of Native or Asian ancestry that could certainly be interpreted as Native, such as Siberian or Beringian.

Does this mean this German/Hungarian man has Native American ancestry?  Of course not, but it does probably mean that the Native population and his ancestral populations did share some genes from the same gene pool thousands of years ago.

While you might think this is improbable, or impossible, consider for a minute that every person outside of Africa today carries some percentage of Neanderthal DNA, and all Europeans also carry Denisovan DNA.  Our DNA does indeed have staying power over the millennia, especially once an entire population or group of people is involved.  We’ve recently seen this same type of scenarios in the full genome sequencing of a 24,000 year old Siberian male skeleton.

Our German/Hungarian man carries 2.4% Neanderthal DNA according to 23andMe and 2.7% according to the Genographic Project, which also reports that he carries 3.9% Denisovan.  The European average is about 2% for Neanderthal.

The net-net of this is that minority admixture is not always what it seems to be, especially when utilizing autosomal DNA to detect small amounts of Native American admixture.  The big picture needs to be taken into consideration.  Caution is advised.

When searching for Native admixture, when possible, both Y DNA and mitochondrial DNA give specific answers for specific pedigree lines relative to ancestry.  Of course, to utilize Y or mtDNA, the tester must descend from the Native ancestor either directly paternally to test the male Y chromosome, or directly matrilineally to test the mitochondrial line.  You can read about this type of testing, and how it works, in my article, Proving Native American Ancestry Using DNA.  You can also read about other ways to prove Native ancestry using autosomal DNA, including how to unravel which pedigree line the Native ancestry descends from, utilizing admixture tools, in the article, “The Autosomal Me.”