My previous post (first Asians were not erectus) concluded with a reference to a paper on Denisovans which stated that: “ modern humans acquired the HLA-B*73 allele in west Asia through admixture with archaic humans called Denisovan” .
Lets dig a little deeper into this HLA allele.
A very uncommon HLA allele
Among all the HLA-B groups (which add up to thirty six ( 36 varieties): B*07, B*08, B*13.... B*83) it is the one with the least quantity of specific HLA halotypes. In other words each of these thirty six groups have their own variations. For instance HLA-B*07 has 124 different allotypes, HLA-B*15 has 221 allotypes.
The one that we are interested in, HLA-B*73 has only two: B*73:01 and B*73:02 (which was detected in only one person in Abi-Rached’s study, so we can set it aside).
Globally its frequency within the human population is quite low, less than 1%, and in most of the world it is closer to zero. But, some places show a very high frequency in comparison to others.
The “raw” data is shown in the following table which combines data from  and :
Frequency / Place – ethnic group
(Data from  unless indicated otherwise. )
2.30% Israel Jews
2.00% United Arab Emirates
1.10% Cameroon Beti
0.90% Georgia Tiblisi,
0.90% Morocco Casablanca
0.90% Burkina Faso Mossi
0.80% Russia Tuva,
0.70% Average for SW Asia 
0.60% Cameroon Bamileke
0.50% Mongolia Khalkha
0.50% Pakistan Pathan
0.50% Iran Baloch
0.50% Tanzanaia Dodoma Kongwa
0.50% Israel Druze
0.40% Average for NE Asia 
0.40% Saudi Arabia
0.30% Bangladesh Dhaka Bangalee
0.30% Jordan Amman
0.30% Uganda Kampala
0.20% Kenya Nandi
0.20% Kenya Luo
0.20% Zimbabwe Harare Shona
0.20% North America, Hispanic 
0.10% Europe 
0.15% Mexico Sonora, Seri 
- We see that it is strongly concentrated in the Middle East and SW Asia, across central and northern Africa and in the Balkans and Caucasus (home of the Dmansi people).
- There are some local “islands” or “singularities”:
- in Mexico (the Seri natives). We will look into them later.
- in Mongolia (close to the Denisovans?)
- in the Russian republic of Tuva (right next to Altai, homeland of the Denisovans)
- The highest frequency is in Pakistan (more later)
- The Israeli Jews and Druze together are second highest.
I love maps! As they show things more clearly than a table. Fortunately, all this data was placed in a map by Abi-Rached . In which he shows the Current distribution “A”. Then, trying to explain the anomalously high Pakistan figure, he defines two models which take into account that the Parsi population now living in Pakistan came from Persia, escaping religious persecution. Figure “B” shows one of these models (they are both very similar) which considers the city of Nishapur as the refugee of the persecuted Zoroastrians. There is yet a third map, “C” which shows the global distribution of the HLA-B*73:01 allele, now including Australia and America, as it comes from another source  it differs slightly from the data given by . Nevertheless, Maps “A” and “C” are very similar and show the same trend.
The maps clearly show some “hot spots” in Mongolia and the Altai Region and an anomaly in Western Mexico (the Seri natives land).
Could the swath across Africa be related to Arab slave trade and possible mixing with locals? Or is it more ancient and shows a returning flow of humans into Africa after acquiring the HLA-B*73 from the Denisovans.
The Balkans, may be due to the Turkish occupation of that area between the 14th and the 19th century.
We see tha China, SE Asia, Australia, Southern Africa and Mots of America and Europe have less than 0.5% frequencies.
So, what does all this mean?
If, as Abi-Rached contends, modern humans picked up the HLA-B*73 allele from Denisovans, they must have “trekked” through Denisovan terrritory. Perhaps the areas where the HLA-B*73 frequency is highest corresponds with the Denisovan homeland: because the humans who settled in the Denisovan territory would have had more time /chances to “mix” with them and breed future generations carrying Denisovan HLA alleles. And from there they would have dispersed to other locations taking the HLA alleles with them.
So, if the maps are correct, Iran is the homeland of the Denisovans. Pakistan’s high figures are just a coincidence, because it is there where the Persian Parsis moved to save their lives.
The Druze and Israeli Jews are also another hot-spot which we will look into later (they are linked to other genetic evidence) , but the high frequency in this region may have two possible explanations:
- It probably indicates that Denisovans were living in that area along the Jordan River valley
- Perhaps it reflects the comings and goings of the Jewish people (remember that they were held captive in Babylon (Babylonian exile) after Nebuchadnezzar forced them in o exile after conquering the kingdom of Judah ca. 590 BC. They probably picked up the “Persian” HLA-B*73 during their captivity there, Persia is right next to Israel. Furthermore, after the creation of the state of Israel, many Jews living in Iraq and Iran were forced to move out and migrate to Israel.
Map “C” clearly shows the hot spot centered in the Middle East and fanning out north west into the area where “late” Denisovan remains were later found (Altai). Suggesting perhaps that the Denisovans occupied this area and later were forced to less favorable lands such as Altai and Southern Central Siberia / Mongolia.
Coincidentally this region (Levant, Caucasus and Altai / South Central Siberia) is where Neanderthal remains have also been found. This may just indicate that it was an area where the last remnants of dwindling homind populations found refuge (i.e. Neanderthals and Denisovans).
Since those (H. heidelbergensi?) who would later become Neanderthals and modern humans split from the Denisovans about 1 million years ago, and Denisovans survived until about 30-50 kya, we can imagine that they lived for 1 million years in this same territory. Let’s put this in context and link it to our previous post.
We had said that about 2 million years ago, H. habilis left Africa, settled in the Caucasus and there evolved (1.75 M years ago) into the Dmanisi people Homo georgicus. These in turn led to H. erectus (which peopled Southeastern Asia and China, moved back to Africa and into Europe). The Dmanisi may have also moved on into America and into Flores Island, where they became the local Hobbits.
All of this happened before 1.5 million years ago. Then there is a blank space of half a million years. The next solid evidence is that 1 million years ago, the Denisovans split from the branch that led to modern humans and Neanderthals: the H. heidelbergensis.
What happened in the Dmanisi homeland during the period 1.5 to 1.0 million years ago? Had all the Dmanisi disappeared, becoming H. erectus and migrating to East Asia, Africa (or America)? Did any remain behind, to become the ancestors of both H. heidelbergensis and the Denisovans? Did they die out?
I have not read about any H. erectus fossils being found in South Western Asia, so if the Dmanisi stayed there, and did not become extinct, they had to become something else.
Getting back to Africa, H. ergaster is said to be the ancestor of H. heidelbergensis and of modern humans, so it is them, the H. ergaster who is the most likely common ancestor of Denisovans and the rest of mankind. And, on the “human - Neanderthal” branch, immediately after the Denisovan split, we can fit H. heidelbergensis, see below (it is a revised version of the tree published in yesterday's post):
This does not tell us anything about the fate of the Dmanisi though. It says that Denisovans split before the appearance of H.heidelbergensis, and marched out of Africa towards the Middle East, settling in South Western Asia where, they would later mix with modern humans and maybe, Neanderthals.
The Seri natives in Mexico
The the Seri language is a linguistic isolate which has no apparent connection with any language in the United States or Mexico. They are only about six hundred Seri left, and a Genetic study  seem to link them to the Warao Indians of Venezuela and suggest that the Seri’s ancestors were part of a first wave of migrants into America.
The paper  however does not mention HLA-B*73.
There is however more data in Abi-Rached’s paper:
Worldwide, ~98% of people carrying B*73 also carry C*15:05
This C*15:05 (also known as Cw*15:05) as seen in the following maps, is also present in Mexico – see red arrow, as well as all the other Asian regions that have a high frequency of B*73. The maps are from  above and  bottom and show the same distribution (with the exception of SE Asia, where there is a spot with very high frequency of C*15:05 in Indochina which is not shown in the bottom map.
We have to understand why the Seri have such a high frequency of B*73 and Mexicans of C*15:05 in comparison to the rest of the American Indians.
That will be the subject of a future post.
 Abi-Rached, et al. (2011). The Shaping of Modern Human Immune Systems by Multiregional Admixture with Archaic Humans Science 25 August 2011: 1209202DOI:10.1126/science.1209202
 Map showing B*7301 dispersion is from This page: Allele frequency net: a database and online repository for immune gene frequencies in worldwide populations. From: Gonzalez-Galarza FF, Christmas S, Middleton D and Jones AR Nucleic Acid Research 2011, 39, D913-D919 and based on Solberg et al (2008) the original is at: http://www.pypop.org/popdata/2008/maps/B-7301.gif
 dbMHC Home, NCBI
 Infante, E., A. Olivo, C. Alaez, F. Williams, D. Middleton, G. de la Rosa, M. J. Pujol, C. Duran, J. L. Navarro, and C. Gorodezky. ( 1999). Molecular analysis of HLA Class I alleles in the Mexican Seri Indians: Implications for their origin. Tissue Antigens 54:35-42.
 Jim Hill, David Yetman, A world revealed by language: a new Seri dictionary and unapologetic speculations on Seri Indian deep history
Patagonian Monsters - Cryptozoology, Myths & legends in Patagonia2011 International Year of Forests Copyright 2009-2011 by Austin Whittall ©