Admixture analyses of North American Indians

In this paper from last year David Reich and his associates analyzed some Amerindian SNP data. Many of the tribes in their dataset showed a significant amount of Caucasoid admixture. Real scientists would have sought to characterize this admixture, but David Reich and his associates simply declared it all to be post-Columbian, and deleted any trace of it from their data before running their analyses. The dataset used for this paper was never made freely accessible.

But this paper from 2007 analyzed many of the same populations using 678 autosomal STRs, and its dataset can be freely downloaded. I have used this dataset here to try to learn more about the Caucasoid admixture in three North American Indian tribes: the Chipewyan, the Cree, and the Ojibwa. The samples for these tribes all came from Canada, and “Ojibwa” is the name more commonly used in Canada for the same tribe that is more often called the Chippewa in the United States. The Cree and the Ojibwa both speak Algonquian languages, while the Chipewyan speak a Na-Dene language.

At the bottom of the post is a plot for a K = 4 admixture analysis of the data from the 2007 paper. I have also included in the analysis the data for Mestizos from this 2008 paper. Below are the mean admixture percentages for the three tribes under consideration. Note that the Caucasoid percentages here are a measure of the Caucasoid admixture beyond the Caucasoid admixture which is found in all Amerindians, and which is a part of the Indianid component.

          Negroid Indianid Mongoloid Caucasoid
Chipewyan    0.20    50.14     27.65     22.01
Cree         0.35    50.47     14.79     34.37
Ojibwa       0.40    52.62     17.11     29.89

Above the K = 4 plot is a plot for a K = 3 analysis that includes the European populations and the two Algonquian tribes. The analysis produces an Algonquian component, a Mediterranean component, and a Nordic component. The mean admixture percentages are below.

          Algonquian Mediterranean Nordic
Orcadian        1.08         35.44  63.48
Russian         2.14          9.25  88.60
Basque          0.57         94.74   4.68
French          1.46         62.30  36.24
Italian         0.76         68.19  31.04
Sardinian       0.72         96.53   2.73
Tuscan          1.44         76.51  22.01
Cree           80.12          9.34  10.54
Ojibwa         90.75          3.30   5.93

The results show that the Caucasoid admixture in the Cree and the Ojibwa is part Nordic and part Mediterranean. The combined Nordic and Mediterranean percentages for the Indians are less than their Caucasoid percentages in the K = 4 analysis, so part of the Caucasoid admixture in the K = 4 analysis becomes part of the Algonquian component in the K = 3 analysis.

Above the K = 3 plot for the Algonquian tribes is a plot for a similar analysis of the Chipewyan. The mean admixture percentages are below.

          Chipewyan Mediterranean Nordic
Orcadian       0.24         20.11  79.66
Russian        0.58          1.93  97.47
Basque         0.30         97.93   1.77
French         0.57         63.57  35.87
Italian        0.32         63.95  35.74
Sardinian      0.22         97.66   2.11
Tuscan         0.62         75.34  24.02
Chipewyan     89.74          1.07   9.18

The results show that the Caucasoid admixture in the Chipewyan is more Nordic than the Caucasoid admixture in the Cree and the Ojibwa. As with the Algonquian component, part of the Caucasoid admixture in the K = 4 analysis becomes part of the Chipewyan component in the K = 3 analysis.

These results can be explained if the Caucasoid admixture in the three tribes is from R1b Nordic males and X Mediterranean females.

There is a lot of evidence that X is a Mediterranean haplogroup. It generally occurs at higher frequencies in Southwest Asia and Southern Europe than in Northern Europe. It has so far not been found in the Nordic hunter-gatherers of Paleolithic and Mesolithic Europe, but it is found in the Mediterranean farmers of Neolithic Europe, and in Europeans generally after the Neolithic. It was found here in samples from Mesolithic and Neolithic Greece, where the Mediterranean haplogroups K, J, and T were also found, but where the quintessentially Nordic haplogroup U was completely absent. In this large study of ancient mtDNA it was completely absent in the Nordic hunter-gatherers, but it was found in eight out of the nine prehistoric German farming cultures studied.

Algonquians have the highest frequencies of both R and X in the Americas. This would explain the presence of both Nordic and Mediterranean admixture in the Cree and the Ojibwa.

One sample of Chipewyan was found to be 62.5% R1, which is lower than the 79.3% found in one sample of Chippewa, but is higher than the 50.5% found in another sample of Chippewa. X is found in the Navajo, but as far as I have been able to determine it is not present in other Na-Dene tribes. The high frequency of R in the Chipewyan, along with a rarity or absence of X would explain the Caucasoid admixture in the Chipewyan being more Nordic than the Caucasoid admixture in Algonquians.

These results have forced me to again change my understanding of Mediterranean migration to Europe. I now think that there were three waves of Mediterraneans from Southwest Asia to Europe: a small wave during the Bølling-Allerød interstadial, a medium wave during the early Holocene and Mesolithic, and a large wave during the Neolithic. I think that the X in North American Indians is from Mediterranean women of the first wave, who along with Nordic men crossed the transatlantic ice bridge that existed during the Younger Dryas.




Posted in Uncategorized

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 )

Google+ photo

You are commenting using your Google+ 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 )

Connecting to %s

%d bloggers like this: