K = 26 admixture analysis of Amerindians and Mestizos

At the bottom of this post is a K = 26 ADMIXTURE plot for HGDP and 1000 Genomes Old World populations, and ancient DNA samples. Above it is an admixture plot for HGDP and 1000 Genomes Amerindian and Mestizo populations based on results from a DIYDodecad calculator created from the ADMIXTURE analysis. Above the calculator plot are the mean admixture proportions for the Amerindian and Mestizo populations.

A while back I had done some low K ADMIXTURE analyses that included the archaic samples, and these analyses would always lump the archaics together with the Negroids. I began to suspect that the Negroid admixture showing up in analyses of ancient samples was actually archaic. This analysis confirms my suspicion. Apparently selection has weeded out a lot of archaic admixture over the past few thousand years.

The Finnish component is modal for La Braña 1, La Braña 2, Ajvide 52, Ajvide 70, and Ire 8.

The Basque component is modal for Gökhem 4, while the Sardinian component is modal for Ötzi and Ste 7. Ste 7 also has some of the Bedouin component and the Japanese component. Ste 7 had the smallest number of SNPs, so the Japanese component may be spurious.

The Gujarati, Yakut, and She components are the largest components for Tianyuan.

There are three distinct patterns for the Caucasoid admixture in Amerindians and Mestizos.

The Colombian, Karitiana, Pima, and Surui Indians have significant amounts of the Finnish and Burusho components, but little to none of the Basque, Sardinian, and Mozabite components.

The Colombian, Mexican, and Puerto Rican Mestizos have significant amounts of the Finnish, Burusho, Basque, Sardinian, and Mozabite components. This is as we would expect, because they have both Amerindian ancestry, from which they would have gotten the Finnish and Burusho components, and Spanish ancestry, from which they would have gotten the Basque, Sardinian, Finnish, and Mozabite components.

The Maya and Peruvians have significant amounts of the Finnish, Burusho, and Basque components, but little to none of the Sardinian and Mozabite components.

The Basque to Sardinian ratio in the Spanish is 2.08. The Colombian, Mexican, and Puerto Rican Mestizos have similar Basque to Sardinian ratios of 2.30, 3.30, and 1.82. But the Peruvians have a much higher Basque to Sardinian ratio of 28.16, and the Basque to Sardinian ratio in the Maya is 1,843.17.

The Basque to Mozabite ratio in the Spanish is 15.23. The Colombian, Mexican, and Puerto Rican Mestizos have similar Basque to Mozabite ratios of 12.96, 15.89, and 9.17. But the Peruvians and Maya have much higher Basque to Mozabite ratios of 47.04 and 26.46.

The Maya and Peruvians are descendants of the people of two high-culture areas of pre-Columbian America. The distinct pattern of Caucasoid admixture in these two peoples is likely the result of admixture from the White Gods, along with a relative lack of admixture from the post-Columbian Spanish.

The MDLP World-22 analysis produced a Cro-Magnon Nordic component (“North-European-Mesolithic”) and an Aryan Nordic component (“North-East-European”). The only people today in which the Cro-Magnon component is modal are Lapps. They have 76.4% of the Cro-Magnon component and 15.5% of the Aryan component.

One of the distinctive characteristics of Cro-Magnon cranial morphology is the short face, as can be seen in the drawing of a Cro-Magnon skull below.

cro-magnon

The same distinctive short face can be seen in these photographs of Lapps (click to enlarge):

lapps-1

lapps-2

The La Braña hunter-gatherers had ~80% of the Cro-Magnon component and none of the Aryan component.

The Oracle results on this page identify northern Finns as being the most similar to the La Braña hunter-gatherers. This makes sense, because northern Finns have the highest amount of the Cro-Magnon component after Lapps.

The Oracle results identify Latvians and Lithuanians as being the most similar to the Swedish hunter-gatherers. Since Latvians and Lithuanians have the highest amount of the Aryan component (~80%), and very little of the Cro-Magnon component (1–4%), it can be inferred that the Swedish hunter-gatherers were Aryan and not Cro-Magnon.

This globe4 analysis showed the Swedish hunter-gatherers as having a spurious 9–13% of the Amerindian component, but showed La Braña 1 as having none of the Amerindian component. This implies that the Nordic admixture in Amerindians is Aryan, not Cro-Magnon.

I now think that Q was originally a proto-Aryan Y haplogroup, and that most, if not all of the Nordic admixture in the available Amerindian samples is from Q males. It’s clear to me now that the haplogroup frequencies and autosomal composition of Amerindians can only be reconciled if Q was originally a proto-Nordic haplogroup. This is especially the case in South America, where the Nordic admixture proportions are about as high as they are further north, but where mt Hg X is nonexistent, and Y Hg R is present at only a low average frequency.

The below distributions of X and R show a decreasing gradient from northeastern North America to southwestern North America, which means that they entered North America from the northeast. The high degree of overlap between the two distributions in North America strongly suggests that the two haplogroups arrived together.

x

r

The Algonquian tribes have the highest frequencies of both X and R in the Americas. Among the Algonquian tribes, the tribe with the highest frequency of R is the Chippewa tribe (79.3% in one sample). The below images of Chippewas show that some of them are the most Caucasoid-looking people in the Americas.

chippewa-1

OLYMPUS DIGITAL CAMERA

chippewa-3

Imacon Color Scanner

Imacon Color Scanner

chippewa-6

chippewa-7

chippewa-8

chippewa-9

chippewa-10

chippewa-11

The two Amerindian subclades of X2 are X2a and X2g, with X2a being the more common of the two. The age of X2a is estimated to be 12,726.6 years. This age coincides with the onset of the Younger Dryas, which lasted from 12,800 to 11,500 years ago.

According to this paper, the R in Amerindians is R1b1a2-M269. This analysis estimated the age of R1b1-L278 as being 15,000 years. If R1b1-L278 originated 15,000 years ago, then it’s reasonable to suppose that R1b1a-P297 originated around 14,000 years ago, and that R1b1a2-M269 originated around 13,000 years ago, just before the Younger Dryas.

The below graph based on Greenland ice core data shows that temperatures during the Younger Dryas were as low as they were during the Last Glacial Maximum.

younger-dryas-graph

The maps below show that the area most severely affected by the Younger Dryas was the North Atlantic region.

younger-dryas-map-1

younger-dryas-map-2

During the Younger Dryas, the transatlantic ice bridge that the Solutreans crossed to reach North America would have reformed, making it possible for the Magdalenians to also reach the New World.

A commenter on this post said that his study of Iroquoian languages had led him to conclude that they are related to Aryan languages. The frequency of Y Hg R in one Iroquois tribe, the Cherokee, is 46.7%.

Below is a map of the word ‘woman’ in the different Algonquian languages. Most of the words contain parts that are similar to the proto-Aryan word for ‘woman’, *gʷénhₐ, or to the derived English word queen.

algonquian

I no longer believe the Kurgan hypothesis. I now think that R1b was present in Western Europe by the time of the Younger Dryas, and that languages ancestral to living and historical Aryan languages were spoken in Western Europe by the time of the Younger Dryas.

According to this tree, the proto-Germanic Y haplogroup is R1b1a2a1a1-S21, and it originated in the middle Bronze Age, which was from 4,000 to 3,500 years ago. But the above-referenced analysis estimated the age of R1b1a2a1a1c-Z381 as 7,000 years. The origination of a subclade of the proto-Germanic clade 7,000 years ago contradicts the Kurgan hypothesis, according to which proto-Aryan had not yet differentiated 7,000 years ago.

The most recent common ancestor of R1b and R1a was estimated to have existed 24,000 years ago, which was around the beginning of the LGM. This suggests that R1 was spread over Europe before the LGM, and that R1b and R1a originated as result of the European population being split into one population that occupied a Western European refuge, and another population that occupied an Eastern European refuge.

I now think that there were no Mediterraneans anywhere in Europe during the Paleolithic. I think that Mediterraneans spread from Southwest Asia along the Mediterranean coast of Europe during the early Holocene and Mesolithic, and that it was these European Mediterraneans, and not Southwest Asian Mediterraneans that introduced agriculture to Northern Europe during the Neolithic. My thinking on this was influenced by the maps on this page, which show that during the Upper Paleolithic no part of Europe was warm enough for long enough to allow for the growth of the type of vegetation that is now found in the Mediterranean area.

            Colombian Indian Karitiana  Maya  Pima Surui
Yakut                  29.08     30.36 26.88 29.52 30.80
Naxi                   17.80     17.88 15.57 18.63 18.43
Japanese               11.86     11.90 10.43 13.99 13.50
Finnish                11.83     10.65 13.12 12.49 10.64
Burusho                10.99     11.00 10.13  9.94  8.94
She                     8.06      8.40  8.60  5.31  6.85
Gujarati                3.38      3.32  2.64  3.07  4.38
Kalash                  2.48      3.01  2.89  2.84  3.03
Basque                  0.97      0.48  5.27  1.38  0.44
Lahu                    0.30      1.08  0.19  0.94  0.59
Brahui                  0.12      0.43  1.07  0.59  0.88
Melanesian              0.91      0.56  0.56  0.40  0.51
Dai                     0.98      0.53  0.15  0.26  0.60
Papuan                  0.15      0.30  0.19  0.35  0.37
Yoruba                  0.36      0.01  0.66  0.12  0.00
Mandenka                0.39      0.00  0.47  0.03  0.00
Kenya-Bantu             0.17      0.01  0.31  0.02  0.01
Druze                   0.00      0.00  0.37  0.00  0.00
Biaka-Pygmy             0.10      0.02  0.17  0.06  0.01
Mozabite                0.00      0.00  0.20  0.00  0.00
Mbuti-Pygmy             0.04      0.02  0.04  0.02  0.01
Archaic                 0.02      0.02  0.03  0.02  0.02
San                     0.00      0.01  0.03  0.01  0.01
Palestinian             0.00      0.00  0.04  0.00  0.00
Bedouin                 0.00      0.00  0.01  0.00  0.00
Sardinian               0.00      0.00  0.00  0.00  0.00

            Colombian Mestizo Mexican Peruvian Puerto Rican
Basque                  28.93   20.48     8.31        29.53
Finnish                 19.45   20.06    19.57        17.49
Yakut                    6.33   12.21    20.63         3.33
Sardinian               12.61    6.20     0.29        16.19
Japanese                 5.17    8.85    13.67         2.82
Naxi                     4.29    8.32    13.90         1.52
Burusho                  2.91    5.34     7.78         1.58
Yoruba                   4.35    2.40     1.37         7.19
Gujarati                 2.00    3.46     5.05         1.71
Mozabite                 2.23    1.29     0.18         3.22
Brahui                   1.94    1.97     0.97         1.47
Kenya-Bantu              1.71    0.89     0.53         3.05
She                      1.03    2.43     2.01         0.62
Kalash                   0.97    1.46     2.16         0.52
Palestinian              0.99    0.74     0.02         2.74
Dai                      0.58    1.02     2.03         0.53
Druze                    1.20    0.62     0.01         1.95
Mandenka                 0.90    0.49     0.23         1.98
Bedouin                  1.04    0.51     0.06         1.29
Biaka-Pygmy              0.38    0.23     0.15         0.35
Lahu                     0.27    0.27     0.36         0.18
Archaic                  0.25    0.24     0.21         0.25
Papuan                   0.17    0.23     0.25         0.14
Melanesian               0.15    0.11     0.14         0.14
San                      0.10    0.09     0.07         0.11
Mbuti-Pygmy              0.06    0.08     0.04         0.09

amer-mest-26

old-world-26

Advertisements
Posted in Uncategorized
11 comments on “K = 26 admixture analysis of Amerindians and Mestizos
  1. alex says:

    My guess is that haplogroup G is the marker of the European Mediterraneans. Haplogroup I is probably Cro Magnon and then R1 Aryan.

  2. genetiker says:

    You may be right about G in Europe being Mediterranean.

    In this analysis Oetzi, who was G2a2a2a-L166, had 6% of the Alpine component.

    But Oetzi didn’t have any of the Alpine component in the MDLP World-22 analysis.

    If G in Europe was Mediterranean, then European Alpines must have been J2.

    There is a very strong relationship between the frequency of J2 and the Alpine component in Europe.

    E1b1b is clearly associated with Mediterraneans and Hamito-Semitic languages.

  3. […] блоггеров (кодовое имя – genetiker) под названием K = 26 admixture analysis of Amerindians and Mestizos. В своей  пространной статье и спорных выводах автор […]

  4. […] on the findings of the K = 26 admixture analysis of Amerindians and Mestizos, it’s safe to assume that the Negroid admixture showing up in the Mal’ta boy is […]

  5. […] the K = 26 admixture analysis of Amerindians and Mestizos post I inferred on the basis of the Oracle results on this page that the hunter-gatherers at Ajvide […]

  6. […] the K = 26 admixture analysis of Amerindians and Mestizos showed, the Negroid admixture in these results is actually […]

  7. […] October 27: K = 26 admixture analysis of Amerindians and Mestizos […]

  8. […] Below are the results of globe4, globe13, and MDLP World-22 analyses of the complete La Braña 1 genome. I’ve also included the results for the calculator that I used in the K = 26 admixture analysis of Amerindians and Mestizos. […]

  9. […] Below are the results of analyzing the Mal’ta and Afontova Gora genomes with the calculator that I used in the K = 26 admixture analysis of Amerindians and Mestizos. […]

  10. […] largest Mongoloid components for pure Amerindians in the K = 26 analysis, in decreasing order, were the Yakut, Naxi, Japanese, and She components. These components appear […]

  11. […] YFull spreadsheet shows that 9 out the 41 1000 Genomes Peruvian samples (22%) are R1b. Note that my K = 26 admixture analysis of Amerindians and Mestizos showed that the 1000 Genomes Peruvian samples have Caucasoid admixture beyond what is found in all […]

Leave a Reply

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

WordPress.com Logo

You are commenting using your WordPress.com 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

Categories
%d bloggers like this: