Analyzing the Native American Clovis Anzick Ancient Results

This ancient DNA truly is the gift that keeps on giving.

Today, Felix Chandrakamur e-mailed me and told me that the Anzick results were not yet fully processed at Gedmatch when I performed a “compare to all.”  He knows this because he knows when he uploaded the results, and after they were finished, he ran the same compare and obtained vastly different results.  I am updating my original article to point to this one, so the data will be accurately reflected.

In fact, the results are utterly fascinating, take your breath away kind of fascinating.  Felix wrote an article about his findings, Clovis-Anzick-1 ancient DNA have matches with living people!

While finding what appear to be contemporary matches for the Anzick child may sound ho-hum, it’s not, and when you look at the results and the message they hold for us, it’s absolutely astounding.

Felix ran his comparison with default values of 7cM.  This is the threshold that is typically utilized as the line in the sand between “real” and IBS, matches – real meaning the results are and could be, if you could find your common ancestor, genealogically relevant.  In this case, that clearly isn’t true.

The exception to this rule is heavily admixed groups, such as Ashkenazi Jewish people who are related to every other Askhenazi Jewish person autosomally.  It seems, looking at these results, that this is the same situation we find with the 12,500 year old Anzick child and currently living people.  This population had to be painfully small for a very long time and the DNA had to exist in every person within that population group for it to be passed in segments this large to people living today.

After receiving Felix’s e-mail, of course, I had to go back and run the compares again.  In particular, I wanted to run the one to many, as he had.

I began at the 1cM level and noticed that I received exactly 1500 results, which seemed to me like a cutoff – not an actual number of matches.  So, I upped that threshold to 2, then 3, then 4, then 5, then 6, then finally to the default of 7.  It was only at 7, the IBS/IBD default, that the results were under the 1500 threshold, at 1466.

1466 current matches?????

This is absolutely amazing.  The Anzick child lived about 12,500 years ago in Montana.  How are 1466 matches to currently living people possible?

Many of these matches are to people from the southwest and Mexico today.  They are not, for the most part, from eastern Canada.

Let’s take a look at what we found.

In the 1466 results, as Felix mentioned, the closest matches match at current “cousin” levels to Anzick.  The highest 7 matches that show haplogroups are haplogroup Q1a3a.  Unfortunately, with the constant renaming of the haplogroups recently, it’s difficult to interpret the haplogroup exactly, which is why we’ve gone to SNP names.  Looking at some of the names and e-mails, several appear to carry Spanish surnames or be from Mexico or South America.

Of the 1466 results:

  • 2 were Y haplogroup C
  • 79 were Y haplogroup Q
  • 520 carried a mitochondrial DNA haplogroup of A, B, C, D, M or X
  • Of the 79 haplogroup Q carriers, 52 also carried a Native mitochondrial haplogroup.
  • A total 549 individuals out of 1466 carried at least one Native American haplogroup, or about 37.5%.  That’s amazingly high.

Of these closest matches who are Y haplogroup Q, they also all carry variant Native American mitochondrial DNA haplogroups as well, so these people may not be heavily admixed.  In other words, they may be almost “pure” Native American.

In order to test this theory, I entered the number of the kit that rated the highest in terms of total cM at 160.1 with the largest segment at 14.8.  You can click on the images to enlarge.


As you can see, this individual is very nearly 100% Native American.

The second individual on the list, who may be from Guatemala, also carries almost no admixture.


Of the highest 21 matches that listed any haplogroup information, all have either or both Native Y or mitochondrial DNA haplogroups.

Out of curiosity, I ran the first person on the list who had neither a Native American Y or mitochondrial haplogroup – both being European.

As you can see, below, they are still clearly heavily Native American, but clearly admixed.


I moved to the last person of the 1466 on this list whose DNA matched at a total of 7cM, who did not carry a Native haplogroup.  This individual, below, is more heavily admixed.

anzick 3.5

Lastly, I ran the same admixture tool on the last person, who had a total of 7cM matching that did have a Native American mitochondrial haplogroup.


Not surprisingly, the individual with almost no non-Native admixture is much more likely to carry the ancient segments in higher percentages than the individuals who are admixed.   This again strongly suggests that at one point, these segments were present in an entire group of Native people and may still be present in very high numbers in people who carry no admixture.

Out of curiosity, and assuming that these first two individuals are not known to be related to each other, I ran them against each other in a one to one comparison.

There were no matches at the default values, but by dropping them just a little, to 5cM and 500 SNPs, they match on 6 segments.


It looks like they should match on chromosome 17 at the 700 SNP/7 cM default threshold.

At 200 SNPs and 2cM, there were 67 segments.  These are clearly ancient in nature and size, but matching just the same.  By lowering the threshold to 100 SNPs and 1cM, they share a whopping 990 segments.

Indeed, these two men very clearly share a lot of population specific DNA from the ancient people of the New World, including that of Anzick male child who lived in Montana 12,500 years ago.

102 thoughts on “Analyzing the Native American Clovis Anzick Ancient Results

    • MRCA means most recent common ancestor and 6 means an estimate of 6 generations. But estimates are just that, estimates and they can be wrong. They are seldom exactly right. I would strongly suggest that you contact Jennifer Zinck at and arrange for a one hour tutorial session so that she can help you learn about all the technical stuff so you can understand what you are seeing, and more importantly, what it means to you.

  1. So I ended up looking into ancient relatives and I am related to Clovis Montana and I used my raw dna data in the ancient calculator to determine how much dna we shared. It said 9.48% My ethinicity does not show native American on any dna sites but my biological grandmother and my birth mother have family records and connections.
    My question is since I share 9.48% with Clovis through dna, would this tie in with my mothers family ethinictiy issue

      • Is there a way to share the screenshot of the percentage of my DNA to his? I’m trying to understand what this even means if I share this.

      • I don’t know much of this how situation and I’m new to DNA but I used the ancient calculator selected Clovis and uploaded my raw data to compare and the total says 9.48%, if you have knowledge about this topic would you please email me so I can send you what I have, if you don’t mind. My email is and anyone can email me about this I’m just as curious lol

      • I went threw gedmatch, used the tool of ancient matches for fun came across Clovis Montana as a high match and I click learn more then used the tool ancient calculator from gedmatch

      • GedMatch shows the threshold at half a cM. Normally, anything less than 7cM is considered probably a match by chance, and not ancestral. Raise the threshold and see what you share. The Ancient Calculator you are referring to is not on GedMatch. You’ve clicked through to another site. I won’t download executable code onto my computer, but I can tell you that to share 9% of your DNA would make you between a first cousin and a first cousin once removed. That’s just not possible. Some people misinterpret decimal points too, so I don’t know what is happening, but I do know that what you are describing just isn’t possible.

      • Ok so how do I find out exactly what I share with him if this isn’t correct. I want to research this but I want to make sure I’m doing this correctly.

      • Compare the Anzick kit number on GedMatch to your kit, doing a one to one comparison. Leave the threshold at the default level. That will tell you if you have a reasonable match. If you don’t have a match at between 5 and 7cM then it’s likely that the matching is by chance. I don’t know what the Ancient Calculator, not found on the GedMatch site, is doing, but guaranteed you don’t share 9% of your DNA with Anzick. That would be between a first cousin and first cousin once removed. Even the closest living match today who is tribally enrolled and from a far northern tribe (or South American) meaning very little admixture doesn’t match that closely. The problem is with the Ancient Calculator. I dropped your threshold level using the kit in your e-mail to 3cM and 300 SNPs, which is extremely low and even at that threshold, you had no matches with Anzick. I’m sorry for this confusion, but I do understand why you were confused.

    • Hi Nicole, I found out that I am related {via gedmatch} to someone who is reported on American Indian research site as having actual clovis anzick 1. (A2e – Clovis (Anzick) specimen)
      The ancient calculator was created by Felix Immanuel, a man who know his stuff and has programs on Gedmatch.
      But there is certainly a bug in that program. It is showing me as 12.9.
      Since I have similar results that are 1.2, that is more likely.
      You can try the program Ancient Ancestry at the same site.
      Great site Roberta, I have learned so much from it!

      • Here is a better explanation from the Ancient DNA creator himself.
        “Felix explains, “The tool checks for small segments chunks within the DNA to see how much they are identical. You need to remove the genealogical perspective and put on the general idea that every human is 99.7% identical. This tool tells you how much of that 0.3 is identical. Instead of SNP by SNP, which mostly match each other by 60 to 70%, it checks smaller chunks of identical segments.”

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s