Announcement

Collapse
No announcement yet.

Clovis-Anzick-1 Amerindian ancient DNA have matches with living people.

Collapse
X
 
  • Filter
  • Time
  • Show
Clear All
new posts

  • #31
    Originally posted by felix View Post
    I just blogged my view on the ancient Clovis-Anzick-1 Amerindian genome and it's matches.

    Link: Ancient Amerindian DNA: How valid are the matches?
    Here are some more mysteries. For kits that match F999912 - Clovis Anzick-1

    A Peruvian kit F349738 is 5.2 generations away from a Mexican.

    A Guatemalan M072434 is less than 6 generations away from multiple Mexicans.

    These Mexicans aren't mixed with Peruvians or Guatemalans. I am familiar with their pedigrees.

    Comment


    • #32
      Originally posted by Armando View Post
      Here are some more mysteries. For kits that match F999912 - Clovis Anzick-1

      A Peruvian kit F349738 is 5.2 generations away from a Mexican.

      A Guatemalan M072434 is less than 6 generations away from multiple Mexicans.

      These Mexicans aren't mixed with Peruvians or Guatemalans. I am familiar with their pedigrees.
      As mentioned in my blog, I can not only verify IBD segments using phasing with the ancient Clovis Anzick-1 DNA sample, but also triangulate using matching segments. I already emailed the authors of the paper regarding these matches and I haven't yet got any response.

      I don't believe the Clovis Anzick-1 sample is 12500 years old. It is just from the last century.

      Comment


      • #33
        Originally posted by felix View Post
        As mentioned in my blog, I can not only verify IBD segments using phasing with the ancient Clovis Anzick-1 DNA sample, but also triangulate using matching segments. I already emailed the authors of the paper regarding these matches and I haven't yet got any response.

        I don't believe the Clovis Anzick-1 sample is 12500 years old. It is just from the last century.
        Yes, I read your blog and I understand what you are saying. I think that you are missing my point though. I'll use a different example.

        The kit F999912 Clovis-Anzick-1 matches F349738 who is from Peru and it shows there is a distance of 3.4 generations. You have shown with phasing that there are "IBD" segments.

        Gedmatch shows a Clovis-Anzick-1 sample matches FN111284 who based on what I can find is an Hispanic from the U.S. with ties to Mexico. If she has ties to Peru I'll have to find another example. In the meantime, assuming I'm correct, Gedmatch shows a distance of 5.3 generations. You have shown with phasing that there are "IBD" segments.

        So 2 of 2 people, 100% success rate, you proved there is IBD. One of them was much further down the list than the other. One is from Peru the other is from North America/Mexico, and the "ancient sample" is from "Montana". There is a very long distance between Peru and North America. If the sample were from the last century how is it that the sample matches two people from very distant places and shows such close relatedness?

        Now on to the triangulation. Clovis Anzick-1, M193252 (likely from northern or central Mexico) and M174237 (from Guatemala). Not as far apart but unlikely to have a common ancestor in the past 7 generations as Gedmatch suggests.

        Triangulation between Clovis Anzick-1, FN111284 (likely Mexican) and A778817 (likely Mexican).

        Based on the above it looks like "IBD" can be older than 500 years/16.6 generations. That's not as big a deal as 12,500 years but it is still not something people like Tim Janzen, Jim Bartlett, and John Walden expect to see after 5 generations.

        What I believe was first stated by John Walden has been repeated by many people on many blogs and forums -

        7 cM matching segment: 30% IBD, 70% IBS
        6 cM matching segment: 20% IBD, 80% IBS
        5 cM matching segment: 5% IBD, 95% IBS
        4 cM matching segment: ca 1% IBD, ca 99% IBS

        According to the ISOGG Wiki 4th cousins once removed, 5.5 generations, the total shared cM is 6.64.

        "Luke Jostins calculated that on average a human has no more then ca. 125 genetic ancestors from the same ancestral generation. This means that only up to the seventh ancestor generation (128 ancestors) segments of those ancestors are detectable in the personal DNA. "

        http://www.isogg.org/wiki/Autosomal_DNA_statistics

        If all of the information from the ISOGG Wiki is correct how can the F999912 Clovis-Anzick-1 match, in less than 6 generations, a Peruvian, a Guatemalan, and multiple Mexicans all at the same time even after phasing and triangulation without the families having intermarried in the recent past?

        Comment


        • #34
          Originally posted by Armando View Post
          What I believe was first stated by John Walden has been repeated by many people on many blogs and forums -

          7 cM matching segment: 30% IBD, 70% IBS
          6 cM matching segment: 20% IBD, 80% IBS
          5 cM matching segment: 5% IBD, 95% IBS
          4 cM matching segment: ca 1% IBD, ca 99% IBS
          IBS does not always mean the segment is 'noise' or to be ignored. There seems to be a wrong perspective that IBS is not required. The difference between IBS and IBD is, an IBD segment has no recombination while an IBS can have. If the IBS does have recombination and yet it may not be noise which is referred as a 'compound segment'. When segments fails in phasing, they aren't noise but rather compound segments due to enogamous marriages / pedigree collapse. An IBS which is a true noise cannot have accidental matches above 2 cM.

          Originally posted by Armando View Post
          "Luke Jostins calculated that on average a human has no more then ca. 125 genetic ancestors from the same ancestral generation. This means that only up to the seventh ancestor generation (128 ancestors) segments of those ancestors are detectable in the personal DNA. "

          http://www.isogg.org/wiki/Autosomal_DNA_statistics

          If all of the information from the ISOGG Wiki is correct how can the F999912 Clovis-Anzick-1 match, in less than 6 generations, a Peruvian, a Guatemalan, and multiple Mexicans all at the same time even after phasing and triangulation without the families having intermarried in the recent past?
          Looks like my DNA even though I know nothing through paper trail

          More often NPEs are not recorded in pedigree as they are not accepted in most societies even today. My paternal grandfather's phased kit matches a European at 7 cM/700+ SNPs who has no South Asian ancestry or admixture. Clearly they are related genetically and could be around 8th cousins, but it could be NPE and there will never be any paper trail - as most Europeans working away from home had double families, one at home and the other at colonies. Most of the matches in FTDNA as close as 3rd cousin is from a different caste. So, I generally give more preference to genetics than what is recorded in pedigree or what is known because DNA don't lie but pedigree can.

          Comment


          • #35
            Originally posted by felix View Post
            IBS does not always mean the segment is 'noise' or to be ignored. There seems to be a wrong perspective that IBS is not required. The difference between IBS and IBD is, an IBD segment has no recombination while an IBS can have. If the IBS does have recombination and yet it may not be noise which is referred as a 'compound segment'. When segments fails in phasing, they aren't noise but rather compound segments due to enogamous marriages / pedigree collapse. An IBS which is a true noise cannot have accidental matches above 2 cM.
            The one thing I haven't seen from the experts is how far back in genealogical time that they have proven IBD be passed on at a segment of 7 cM.

            At least you are using phased data, large segments, and triangulation. That is something a lot of other people are not using in trying to find a connection.

            Originally posted by felix View Post

            Looks like my DNA even though I know nothing through paper trail

            More often NPEs are not recorded in pedigree as they are not accepted in most societies even today. My paternal grandfather's phased kit matches a European at 7 cM/700+ SNPs who has no South Asian ancestry or admixture. Clearly they are related genetically and could be around 8th cousins, but it could be NPE and there will never be any paper trail - as most Europeans working away from home had double families, one at home and the other at colonies. Most of the matches in FTDNA as close as 3rd cousin is from a different caste. So, I generally give more preference to genetics than what is recorded in pedigree or what is known because DNA don't lie but pedigree can.
            I am fully aware of NPEs and I do understand the history of the colonies in Latin America and South Asia and I took that into consideration. In your case it is likely a European ancestor that ties the two of you together. The British were in South Asia making it to where distance is not a barrier.

            The geographical distance and supposed genealogical distances make an NPE a very unlikely possibility. If we were talking about a timescale of 500 years I could agree it is a slight possibility.

            There is such a little chance that in recent history a Mexican went to Peru and even less that a Peruvian went to Mexico and Guatemala it is basically an impossibility that the three regions are tied to an NPE especially since what ties them to together is Native American ancestry and not Spanish ancestry. If were only talking about Mexicans this distances would not be an issue. Colonial Mexico was controlled by Spain and not by Peru. Colonial Peru was controlled by Spain and not by Mexico. Even though movement between Mexico and Peru did happen it was minor.

            I got confirmation that M193252 and A778817 are from Mexico.

            I also got confirmation that F15295 is from Mexico. He matches F999912 Clovis-Anzick-1 and he matches kit F349738 from Peru as well as phased kit PF338998P1.

            Some of the people from Mexico, not mentioned above, are from northern Mexico and do not have ancestry in the ports of Mexico and have a well documented tree. Cuckoldry isn't as common as you might think. Most children from unwed parents in Mexico are called hijos naturales when the mother is named or hijo de padres no conocidos when the child is abandoned. Genealogical records in South Asia are not the same as genealogical records in Mexico which have some of the best records in the world.

            This mystery of the Peruvian, the Guatemalans, and the Mexicans being related to 999912 Clovis-Anzick-1 is not due to an NPE.

            Comment


            • #36
              Originally posted by felix View Post
              [----] I don't believe the Clovis Anzick-1 sample is 12500 years old. It is just from the last century.
              Contamination ? Like with the Denisovan DNA ?

              Comment


              • #37
                Originally posted by dna View Post
                Contamination ? Like with the Denisovan DNA ?
                That does not explain the matches from different regions of Mexico, Guatemala, and Peru at 7cM and 700 SNPs and neither does an NPE.

                How likely is it that a person mixed with Mexican, Guatemalan, and Peruvian ancestry contaminated the specimen and we were lucky enough to have enough of his relatives to have tested and show to match with 7cM and 700 SNPs?

                Comment


                • #38
                  Roberta Estes has a new blog post on this subject

                  http://dna-explained.com/2014/09/25/...-do-they-mean/

                  Comment


                  • #39
                    These results (in Roberta Estes' words "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 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.

                    Comment


                    • #40
                      Originally posted by Armando View Post
                      Roberta Estes has a new blog post on this subject

                      http://dna-explained.com/2014/09/25/...-do-they-mean/
                      Thank you Armando.

                      So we had an estimate that the founder population was no more than 80, and now number 10 is being tossed around. Let's wait until somebody clever in mathematics and populations genetics (hmm, is there a difference between these two specialties ?) calculates that for us

                      P.S. I hope the picture in the blog is not about anything DNA related, since they clearly have no gloves

                      Comment


                      • #41
                        Originally posted by Ann Turner View Post
                        These results (in Roberta Estes' words "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 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.
                        Hi Ann, I was also able to triangulate kits M162053, A418973, M161504 with each other and Clovis Anzick-1 sample (F999912) for a segment longer than 11 cM / 900+ SNPs. I had also mentioned more triangulations in my other blog.

                        Comment


                        • #42
                          Felix, could you run some sample data through David Pike's utility,but change this parameter to "No"

                          "Treat No-Calls as Matching SNPs?"

                          http://www.math.mun.ca/~dapike/FF23utils/pair-comp.php

                          Comment


                          • #43
                            Originally posted by Ann Turner View Post
                            Felix, could you run some sample data through David Pike's utility,but change this parameter to "No"

                            "Treat No-Calls as Matching SNPs?"

                            http://www.math.mun.ca/~dapike/FF23utils/pair-comp.php
                            Hi Ann, The problem is, I don't have access to autosomal files who have significant Amerindian DNA.

                            But Comparing Kit F999912 (Clovis Anzick-1) and A418973 (Juan Jhong) and many others with threshold 7 cM and 700+ SNPs, with Mismatch Evaluation Window and Mismatch-bunching Limit - both set to 700 seem to provide the below info. So, there are no error SNPs but not sure of no-calls.


                            Chr Start Location End Location Centimorgans (cM) SNPs
                            5 121480057 130878837 7.9 713
                            22 17269490 24752980 16.7 1076
                            Last edited by felix; 25th September 2014, 08:42 PM.

                            Comment


                            • #44
                              Originally posted by Ann Turner View Post
                              These results (in Roberta Estes' words "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 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.
                              Gedmatch is causing people that do not have a common ancestor in genealogical time make it seem as though they do even for living people.

                              For instance Kit F349738 is from a remote part of Peru. His 3rd match in 'One-to-many' matches is 10.7 cM for total and largest cM and 700 SNPs and provides a GD of 5.2 generations. That match to F349738 from Peru does not have any ancestors other than from Mexico since 1750 which has been corroborated by the town historian that has access to many documents, family trees, and DNA results of many people from the town.

                              These two probably aren't related even in the past 500 years.

                              Comment


                              • #45
                                Originally posted by felix View Post
                                Hi Ann, The problem is, I don't have access to autosomal files who have significant Amerindian DNA.

                                But Comparing Kit F999912 (Clovis Anzick-1) and A418973 (Juan Jhong) and many others with threshold 7 cM and 700+ SNPs, with Mismatch Evaluation Window and Mismatch-bunching Limit - both set to 700 seem to provide the below info. So, there are no error SNPs but not sure of no-calls.


                                Chr Start Location End Location Centimorgans (cM) SNPs
                                5 121480057 130878837 7.9 713
                                22 17269490 24752980 16.7 1076
                                Moreno et al states in their study that they are willing to share samples that they have under certain circumstances.

                                http://www.sciencemag.org/content/34.../1280.abstract

                                Comment

                                Working...
                                X