jump to navigation

28,000 Year Old Cro-Magnon mtDNA More Modern Than Anything Else March 10, 2009

Posted by ethnicgenome in Caucasians.
Tags: , , , ,
1 comment so far

This is very interesting.

The open access journal PLoS One has published a new ancient DNA paper, “A 28,000 Years Old Cro-Magnon mtDNA Sequence Differs from All Potentially Contaminating Modern Sequences.” One which establishes, yet again, that the Neandertal mitochondrial hypervariable region I is much different from modern humans. This was done by extracting mtDNA from the Paglicci 23 specimen, understood to be a Cro-Magnon individual.

Paglicci 23 was discovered by Francesco Mallegni, one of the authors of the current paper, in 2003 from the Paglicci Cave site. I think the Paglicci Cave site is currently the earliest Aurignacian-Gravettian site in Italy — dating to be 34,000 years old. Anyone know for sure?

Either way, the Paglicci 23 and is represented by a tibia, skulls, jaw and maxilla. Radiocarbon dating indicates the remains are 28,100 years old. It was not excavated nor handled in sterile conditions. It was washed and analyzed by seven people since then, raising the possibility of contamination from modern human DNA.

In 2005, a piece of the tibia and two pieces of the skull were moved to a DNA lab at the University of Florence, where DNA was extracted, the hypervariable region I of the mtDNA amplified and sequenced. It should be noted that the bone was washed with bleach and irradiated with UV light and mtDNA was isolated from tissue inside the bone — not directly handled by people. The mtDNA of the people who handled the remains was also isolated and sequenced. This was done to see if their DNA contaminated the sample, since the remains weren’t handled under sterile conditions.

This procedure was repeated in another laboratory, by Carles Lalueza Fox in Barcelona, using a tibia fragment. Carles Lalueza Fox is a big name in ancient DNA studies, one of the guys who sequenced FOXP2 from Neandertal remains. Oh also, using Neandertal specific primers for the PCR, yielded no results — indicating that the remains were modern human… but more on that later.

Anne Holden blogger at 23andMe, who says she specializes in science writing, hails this paper in providing a “novel way,” of identifying contamination –

“by analyzing the DNA of everyone who touched a fossil for comparison in order to rule out contamination.”

She’s mistaken. What Caramelli et al. have done in this paper is hardly novel. I covered news of a May, 2008 paper in Science which isolated and extract mtDNA from 4,000 year old hair from Greenland. The hair was also not excavated under sterile conditions. In that paper, the DNA of everyone who handled the sample was also screen and compared to the sample in order to rule out contamination. And that’s just one example of preventative measures in sequencing ancient DNA that immediately comes to mind.

Anyways, the relationship between the Paglicci 23 sequence and all the sequences from the seven individuals who touched the specimen was investigated. In addition, the sequences were compared against other Cro-Magnon and Neandertal sequences. They had some problems which were resolved by treating the DNA extract was treated with Uracyl-N-Glycosidase (UNG). The authors conclude that,

“the Paglicci 23 individual carried a mtDNA sequence that is still common in Europe, and which radically differs from those of the almost contemporary Neandertals, demonstrating a genealogical continuity across 28,000 years, from Cro-Magnoid to modern Europeans.”

In fact, they write,

“we concluded that the sequence obtained from the Paglicci 23 specimen is the CRS.”

The press is saying that the results are first ever demonstration ‘that the anatomical differences between Neandertals and Cro-Magnoids were associated with clear genetic differences.’ That’s overshooting the analysis. There are clear differences in the mitochondrial genome of Neandertals and Homo sapiens, and this is not the first paper to document it. And all this shows is that the Neandertals are not the ancestors of modern Europeans.

Caramelli, D., Milani, L., Vai, S., Modi, A., Pecchioli, E., Girardi, M., Pilli, E., Lari, M., Lippi, B., Ronchitelli, A., Mallegni, F., Casoli, A., Bertorelle, G., Barbujani, G., Harpending, H. (2008). A 28,000 Years Old Cro-Magnon mtDNA Sequence Differs from All Potentially Contaminating Modern Sequences. PLoS ONE, 3(7), e2700. DOI: 10.1371/journal.pone.0002700


Race vs. Ethnicity in DNA March 9, 2009

Posted by ethnicgenome in Genetics (General).
Tags: , , ,

What is the difference between a race and an ethnicity?

The differences between races are the haplogroups themselves.

Races belong to different haplogroups 85-95% of the time, with remaining being the inevitable outside gene flow. (See this article on why being in a different haplogroup does not mean that you are genetically different from most people in your ethnicity.)

But ethnicities differ from one another based on further, much more recent splits of the same haplogroup.

Haplogroup J is common in Europe and the Near East. This is true for both mtDNA and Y-DNA, though the letter J assigned here is just a coincidence and not based on any similarity of J in mtDNA to J in Y-DNA.

In mtDNA, J1 is spread around Europe, while J2 is mostly concentrated around the Mediterranean. In Y-DNA, J1 is largely an Arab haplogroup and J2 is also concentrated around the Mediterranean, just like J2 mtDNA.

Looking at it deeper, the vast majority of the British people who belong to J haplogroup (mtDNA) are J1 and not J2.

Digging deeper, J16172 and J18192 are relatively common among the Scots, but almost never among the Anglo-Saxons and other continental Europeans (though it is present in smaller numbers in Scandinavia).

Any ethnicity that is overwhelmingly part of J and/or other Caucasian haplogroups, would have to be classified as part of the white race.

The next level would split the race into subgroups (Nordic, Mediterranean, and so on).

As we look deeper, sometimes several levels deeper, we begin to reach ethnicities. More and more similarities will get you closer to being related: distant relatives (third cousins), close relatives (first cousins) and nuclear family (siblings).

The idea that there are no races would have to therefore logically lead one to conclude that there are no families.

If race is a social construct, then so too must be the family.

mtDNA in Great Britain March 9, 2009

Posted by ethnicgenome in United Kingdom.
Tags: , , , , , , , , ,
1 comment so far

British mtDNA haplogroups are:

H – 55
J – 15
T – 7
U – 7
K – 3

The predominance of H haplogroup (mtDNA) is common around Europe and the Near East. But J is more common on the British Isles than elsewhere. It is generally about 2% of the European population. Among the British (the English, the Scots, the Welch and the Irish), there is a much higher amount of J.

J is also common in the Near East in similar numbers as in Britain.

As with all haplogroups, it is further split. I will go down one level only to avoid confusing everyone. J1 is spread around Europe, while J2 is mostly concentrated around the Mediterranean. (This is not to be confused with Y-DNA haplogroups by the same name where J2 is also concentrated around the Mediterranean, but J1 is largely an Arab haplogroup.)

All the British ethnicities are overwhelmingly J1 and not J2.

There are some differences however between the native Britons and the Anglo-Saxons who invaded from German regions near Netherlands. British Celts do show some distinction from continental Europeans, while the English show a similarity to West Europeans.

The distinctions here are within the haplogroups.

These are obviously not racial differences, but rather ethnic ones.

Prejudice may be the best way to determine genetic history March 9, 2009

Posted by ethnicgenome in Genetics (General).
Tags: , , ,

I have a strong feeling that people knowing your ethnicity’s haplogroups will tell you more than your own. Y-DNA and mtDNA are a very small part of one’s genetic history. Your mother’s father and all his ancestors, and of your father’s mother and all her ancestors can’t be seen. Same with your maternal grandmother’s father and with your paternal grandfather’s mother. And so on, so forth.

But for a nation, we have too big a sample. If 20% of a given ethnic group are in haplogroup K, my guess would be that roughly roughly 20% of the genes of all the people in that ethnic group are K. So although your DNA test may show that you may be haplogroup K and your neighbor is haplogroup H, the reality is that you both likely have a similar background.

This is just a guess of course, but I am fairly certain about it considering how much mixing there has been among people of the same nation over thousands of years.

The other reason I think so is that people of the same nation look mostly somewhat similar. Not identical of course, but usually similar enough that they can say things like, “this guy looks Swedish and this guy looks Italian.”

Again people may not necessarily look like their their own ethnicity (an Italian may have a Swedish great grandfather whom he never met, but whom he resembles), but there’s a particular “look” to most ethnicities. The closer they are, the harder it is to tell. Thus the Poles and the Russians may be difficult to tell apart sometimes (except usually by the Russians and the Poles themselves), but it is much easier to tell the Poles from the Greeks (again not always, but usually).

The fact that there’s an “Irish look” or a “Dutch look” leads me to believe that most people within a particular ethnicity same a similar amount of each haplogroup. Otherwise, they would look much more different.

There are of course countries where people seem to have different amounts of their haplogroups. Italy, Israel, Turkey, Russia, Finland would be a few examples. But now you can see a difference between people in these cases (north Italy and Finland vs. South; Eastern Russia vs. Western; Sephardic vs. Ashkenazi Jews).

Even though both Sephardic and Askhenazi Jews descended from the same haplogroups (mtDNA and Y-DNA), they have a different mix and therefore look different.

I would therefore guess that it is not that 19% of the Ashkenazim are J1 (usually Arabs), 23% are J2 (usually Mediterranean Europeans), 13% R1a (Slavs, Aryans), 11.5% are R1b (Western Europe), etc. Instead, the Ashkenazim all have Y-DNA somewhere around these percentages.

Likewise, I believe that the people native to northern parts of Finland and eastern parts of Russia (beyond Ural) are probably mostly of Asian descent, genetically linked to some Amerindian tribes. It’s not that 60% of the people there are part of haplogroup N (Y-DNA), but rather almost all the people there are 40-80% N.

Some may have more of a given haplogroup and some may have less, but I would bet that estimating by using “prejudice” based on a person’s ethnicity is more accurate, if not completely exact, than testing one’s mtDNA and Y-DNA, and leaving out all other ancestry.

European Genetic Admixtures: Y-DNA March 5, 2009

Posted by ethnicgenome in Caucasians.
Tags: , , , , , , , ,

This post is a continuation of “Mapping out the borders of whiteness“. If you haven’t read it, you probably should to understand this more clearly here.

The following classifications aren’t hard facts. A scientist would frankly laugh at them because all of these haplogroups appear among all European and Near Eastern peoples.

For example, R1b is generally thought of as West European where 60-90% of people belong to this haplogroup. In Central Europe (Germany, Czech Republic, continental Italy) the frequency is about 40%.

But we see the same ~40% frequency of R1b among Armenians, Ossetians and Jordanians in the Dead Sea region, which is higher than among Swedes, Norwegians and Sicilians.

It also present in significant numbers throughout the Near East (from 4% in UAE to 16% in Turkey) and Eastern Europe (from 2% among Ukrainians to 15% among Bulgarians).

As I’ve said before, haplogroups can be split further and are somewhat different from region to region. But the differences aren’t radically different and differences within a haplogroup are frankly still small enough to be placed in the same haplogroup.

With that said, you can classify certain haplogroups as being most common in particular regions. That is what the following will attempt to do.


R1b: Western Europe (Celtic, Basque, Italic, Frisian, Saxon)
R1a: Eastern Europe (Slavic, Aryan)
I1: Northern Europe (Germanic, Scandinavian)
I2a: Southern Slavic (Balkans and southwestern Ukraine. I2a1 is also present in large numbers in Sardinia, Basques.)
I2b: Western Europe (Germany, Denmark, Netherlands, Belgium and England)


J1: Arabs, Jews.
J2: Spanish, Italian, Sicilian, Greek, Anatolian Turkish, Jewish, and some French and Portuguese.
T: Small haplogroup that is most present among Egyptians, Iraqis, Serbs and the Spanish. Despite small presence anywhere and especially in Europe, it was the haplogroup of President Thomas Jefferson, Russia’s Czar Nicholas II and several English royals.
G: Most present in Caucasus (southern Russia), which is culturally Middle Eastern. About 60% of Ossetians; 30% of Georgians, Kabardinians and Balkarians.
E1b1b: North Africa, Jews and southern Europe in the same regions as J2, but especially Greece and Serbia at 27% and 24%. (Most of E haplogroup migrated back from the Near East to Africa, but E1b1b stayed above Sahara and spread into Europe. It is now the most common of Mediterranean haplogroups among most European ethnicities.)


N: First appeared in Southeast Asia. Its highest frequency occurs among the Finnic and Baltic peoples of northern and eastern Europe, the Ob-Ugric and Northern Samoyedic peoples of western Siberia, and the Siberian Turkic-speaking Yakuts.

Q: It is the dominant haplogroup among Native Americans. Also highly present among some Siberian Mongoloids like the Kets. (Their language has been linked to the Na-Dene languages among Indians in Western Canada and Alaska.)

European populations have only small traces of sub-Saharan haplogroups, so I will not include them here.

For simplicity’s sake, let’s combine haplogroups into European, Mediterranean/Middle Eastern and Asian/Amerindian.

The most European people are, not surprisingly, those on the British islands, particularly the Scots. They are 95% European, 4.5% Mediterranean and 0.5% non-white. The English are 92% – 7.5% – 0.5%. The Welch are 91.5% – 8.5% – 0%.

A little more surprising is the near complete lack of non-European genes among the Poles who are 94.5% – 5.5% – 0%.

This is somewhat surprising because the people with the most non-Caucasian Y-DNA are those around Poland. Russia is 67.5% – 8% – 24.5%. Estonia is 58.5% – 7.5% – 34.5%. Latvia is 59% – 1.5% – 39.5%. Lithuania is 56% – 1.5% – 42.5%.

And the Finns are the only people in Europe or the Near East who are mostly non-Caucasian at 40% – 1% – 59%. In fact, the Finns have so much non-Caucasian DNA that they could be classified as non-white at first look. But as I explained, their non-white Y-DNA is located mostly in the northern half of the country, so the south should be regarded as white and the north as mixed race.

Sweden has a lot less non-Caucasian Y-DNA at 88.5% – 2.5% – 7.5%.

The most Mediterranean/Middle Eastern people are the Greeks at 39.5% – 60.5% – 0%. Italy is 58% – 42% – 0 %. We also see a strong Mediterranean/Middle Eastern genetic influence in the Balkans and southeast Europe. Serbia is 57% – 43% – 0% and Hungary is 65.5% – 31% – 3.5%, while the Muslim Albania is 48% – 52% – 0%.

Interestingly, Portugal has a lot more J2 and E1b1b than Spain (but on the mtDNA side, Spain has the most non-Caucausian DNA at 23%). Portugal is 61.5% – 38.5% – 0%, while Spain is 79% – 14% – 0%.

Similarly interesting is the much higher presence of Mediterranean/Middle Eastern haplogroups among the Austrians (69% – 30% – 1%) than among the Germans (85.5% – 13.5% – 1%). The French are somewhere between them at 80% – 20% – 0%.

* I will do another post soon on the mtDNA, which is a bit more complicated for Caucasians than Y-DNA.

Mapping out the borders of whiteness March 3, 2009

Posted by ethnicgenome in Caucasians.
Tags: , , , , , ,

DNA studies have permitted to categorize all humans on Earth in genealogical groups sharing one common ancestor at one given point in prehistory. They are called haplogroups. There are two kinds of haplogroups: the paternally inherited Y-chromosome DNA (Y-DNA) haplogroups, and the maternally inherited mitochondrial DNA (mtDNA) haplogroups. They respectively indicate the agnatic (or patrilineal) and cognatic (or matrilineal) ancestry.

Haplogroups are divided with different letters, and then with numbers. Thus, we have haplogroup I, which is then further subdivided into I1 and I2. They could be (and are) further divided to achieve more specific definitions and you can arrive at something like Q1a3a1 (Tinuna and Wayuu Amerindians).

In this blog I will try to define the “borders of whiteness”, where the white race begins and ends geographically and which populations should be classified as Caucasian.

Let’s begin by saying that religion should not be used as a guide on race. Chechens, for instance, were once Christian but are Muslim today. When they converted from Christianity to Islam, they did not change their DNA, just their beliefs. Similarly, when Indians in Mexico converted to Catholicism, they did not become genetically similar to the Poles.

Arguments have also been made that only those originally from Europe could be classified as white. The argument is false, genetically-speaking.

The same mtDNA haplogroups (Six Daughters of Eve that derived from haplogroup R as well as haplogroups X, I and W) that are present in Europe are also present in the Near East.

In Europe and the Near East, mtDNA haplogroups are quite evenly spread over the continent, and therefore cannot be associated easily with ancient ethnicities.

It is impossible to decide to classify a person’s race based on arbitrary geographic borders of a continent or a country. Two people in the same mtDNA and Y-DNA haplogroups must be part of the same race.

For that reason, Europe and the Near East must be part of the same race.

Even the distribution of haplogroups is very similar. Among just about everyone in Europe and the Near East, H is more common than U which is more common than J. (One major exception are East European Jews, about a third of whom belong to haplogroup K, which normally is highest in western Europe and British Isles, but isn’t present in such high numbers among those populations, indicating a recent, West European, female founder. Otherwise, everyone else I’ve researched shows very similar distribution patterns.)

The most common haplogroup among Europe is H, with about 40% of the population. About a quarter of Near Eastern Arabs are part of haplogroup H and the Jews are between the Arabs and the Europeans (about 40% of Russian Jews and 33% of Romanian Jews are in haplogroup H or the closely related HV).

People of different races (Pacific Islanders, for instance) belong to completely different haplogroups.

The distribution is somewhat different when it comes to Y-chromosome DNA and not as evenly spread.

Haplogroup I is in northern Europe as well as in the Balkans, R is in the middle of Europe and J stretches from western Portugal and southern Spain along all of southern Europe and into Israel and the Arab states.

It has been argued by some amateur “racialist” geneticists that J1 haplogroup should be classified as non-white because while it is very common among the Arabs, it is not common among the Europeans and somewhere in between among the Jews. (Ashkenazim are 19% J1 and 23% J2;. Sephardim are 12% J1 and 29% J2, but also 29.5% western European R1b.)

This is false logic, however, based on the idea that arbitrary geographic definitions determine genetics.

J2 Europeans (many Greeks, Italians, Spanish and Portugese, as well as southern French) are closer to J1 Arabs than to R1a Europeans, so classifying J2 with R1a as one race, and J1 as a separate race makes no sense at all. It makes even less sense to classify as separate races a J2 Italian and a J2 Lebanese just because they are on different continents.

Whatever differences may exist between J1 and J2, they pale in comparison to real racial differences with Orientals, Amerindians and sub-Saharan Africans. For people to be part of another race, at the very least they shouldn’t be part of the same haplogroup.

Also, the southern J haplogroup and the northern I haplogroup both descended from IJ and are therefore closely related, closer in fact than either is to haplogroup R.

Arabs, therefore, should not be defined as non-white due to their Y-DNA. Instead, they should be classified as a mixed (Caucasoid, Mongoloid and Negroid) race of people due to the high influx of African and (east and south) Asian DNA.

About 20% of Arab mtDNA comes from Africa and another 18% from the East, with only 62% being white. With so much outside DNA, I would classify the Arabs as the first “border” population, a mixed race between whites and non-whites.

(The so-called Arabs in northern Africa are a different issue. They belong mostly (50-80%) to the E haplogroup, which is indigenous to the region. Being Arab seems to be more a matter of tradition than genetics for the North Africans.

Sephardim, on the other hand, are white despite theories pushed on Stormfront. A mix of mostly J2 and R1b is white whether we are talking about Spaniards or Sephardim.

Classifying Sephardim as non-white and the Spanish as white also flies in the face of genetic studies that show that 23% of Spanish mtDNA is non-white, more than among the Sephardim.)

In the North, the Asian N haplogroup is present among 58.5% of Finns, 42% of Lithuanians, 38% of Latvians and 34% of Estonians.

About 23% of Russians are also in haplogroup N. However, these 23% aren’t spread evenly among the whole Russian population.

Just as Turks from the east of their country have more Asian blood than those from the west, the same is true for Russians. Russians west of Ural, where R1a haplogroup is largest, should no doubt be classified as white in most cases. Those east of Ural, on the other hand, are mostly a mixed race Caucasoid-Mongoloid people, more mixed in some regions than others.

Likewise in Finland, those in the south of the country are Caucasian (I1 haplogroup), while those in the north are a Caucasoid-Mongoloid mix.

Again we see that drawing the genetic line along Europe’s borders makes no sense. Northern Finland and northwestern Russia are in Europe, yet the populations there are racially mixed.

So looking at the genetic map, I would classify the “borders of whiteness” from northern half of Finland stretching along northern Russia, going down along eastern Russia in Siberia and the Far East, stretching back southwest into to eastern Turkey and the Arab states. These people (northern Finns, northern Russians, eastern Russians, eastern Turks and the Arabs) are predominantly a mix of whites and others. On one side of them are whites. On the other are non-whites (sub-Saharan Africans, Central Asian Orientals, and so on).

All the people inside the map I just drew would have to be classified as white, including not just Europeans but also Anatolian Turks, Jews (Ashkenazi and Sephardi), Kurds, Dagestanis (who are technically in Europe, are are culturally Middle Eastern) and others.

mtDNA: ‘Caucasian Eve’ Had Only Six Daughters March 1, 2009

Posted by ethnicgenome in Caucasians.
Tags: , , , , , ,

Over 90% of all Caucasoids – people native to Europe, North Africa and the Middle East – are descendants of the same 6 women who lived 20,000 to 30,000 years ago.

Their mtDNA (maternal line) haplogroups are H, J, K, T, V and U. All of them came from haplogroup R, which originated somewhere between the Black Sea and the Caspian Sea. That region is known as Caucasus, which is why white people are referred to as Caucasians.

East Asian haplogroups B and F, as well as the Pacific haplogroup P also derived from R, but due to the vast geographic separation from the above 6 haplogroups, they became vastly different.

There’re also some white people belonging to haplogroup X (and a small number of I and W), which is why Bryan Sykes referred to Caucasoid mtDNA as having descended from the 7 daughters of Eve.

But as I explained, I don’t believe it to be accurate because X is a near-universal haplogroup that is found in small numbers throughout much of the world, including among Amerindians, Pakistanis and others. It is therefore more an international haplogroup than a Caucasian one.

X originated in Iran or Turkmenistan, and not in the Caucasus. It is part of the haplogroup IWX and is related to haplotype R only because both IWX and R descended from haplotype N 30,000 to 55,000 years ago.

There are similarly 9 women who are the ancestors of all Japanese people, and 4 women (haplogroups A2, B2, C1 and D1) are the ancestors of nearly all American Indians (and most of the remaining are descendants of haplogroup X).

This how mtDNA developed since the first humans left Africa. (You have to adjust your eyes because this map has Africa on top.)

mtdna haplogroups

Haplogroup X: The International mtDNA March 1, 2009

Posted by ethnicgenome in Genetics (General).
Tags: , , , ,

Probably the most interesting mtDNA (maternal line) haplogroup X which has widespread global distribution without major regions of distinct localization. It is present in Europe, Asia, North Africa and the Americas. (This is not due to recent immigration, but among native populations. Thus when discuss the Americas, I am not talking about the English, the Spanish and others who settled it in recent centuries, but rather the indigenous American Indian tribes.)

For example, more than 90% of Caucausians are part of 6 haplogroups that derived from haplogroup R, but 2% are part of haplogroup X. Similarly, more than 95% of American Indians derived from 6 haplogroups of East Asian origin, but 3% of them are part of haplogroup X.

The founders of X haplogroup were most likely Druze because they have a particularly large percentage of their population (26%) belonging to this haplogroup, indicating the founder effect. (No ethnicity belongs exclusively to haplogroup and 26% is actually rather high, especially for this particular haplogroup.)

The X haplogroup is the only Amerindian haplogroup that does not show a strong connection to East Asia and is particularly common among the Sioux (15%), the Nuu-Chah-Nulth (11%–13%), the Navajo (7%), and the Yakima (5%).

Among whites, In the nation of Georgia, about 8% belong to this group. There’s also a higher than average presence of this haplogroup in Greece, Siberia and parts of Scotland. All native European populations have at least 0.5% belonging to this group among their native populations.

The only population that presently resides in Europe (that I found) which had no members of the X haplogroup were Russian Ashkenazi Jews, according to Nature.

It is rather curious considering that non-Jewish Russians have a higher than average percentage (3.5%) of X haplogroup as do the the people around Israel, including the Druze and the Turks, while the Palestinians and the Egyptians have about average percentage of X haplogroup.