2Haplogroup G Project
DNA Results and Groupings
G2a1a persons with origins in the Caucasus Mountains region
The most characteristic value for G2a1a men is 10 for marker DYS392, So far all known G2a1a persons
have theis value. In available samples, G2a1a men account for most of the G persons with
a value of 10 for DYS392. So the finding alone cannot predict who is G2a1a, but the presence of a value of 10
should raise a red flag. Of lesser importance, all known G2a1a persons also have a value of 12 at marker
DYS425, a slowly mutating marker that has some variability. In addition, researcher Thomas Krahn has
suggested that marker DYF411 may have some value in identifying G2a1a persons. DYS392 is available
in the basic marker panel offered by most genealogy DNA labs. DYS 425 is a little more difficult to obtain, but most
of the samples available to us with this value are in markers 38-67 of a panel of tests offered by Family Tree DNA.
Marker DYF411 is a special order item, and not enough data are currently available to offer any useful
information about it.
In addition to the characteristic marker values mentioned, some G2a1a persons can share some oddities in the
marker values, and all will have a relatively close genetic distance to known members of the G2a1a
group. A close genetic distance will be exemplified in relatively few mismatches between two comparied samples.
In addition, we have used special DNA software (Fluxus) to produce phylogenetic diagrams using
67-marker samples. Using 67 markers, these diagrams, which give extra weight to the slower-
mutating markers, have been relatively successful in showing correct relationships. The clustering of men on the
diagrams because of closer relationships allows us to identify persons who are likely G2a1a but lack special
testing for G2a1a. The diagrams also suggest who should compose a subgroup due to their clustered presence
on a distinct separate branch on the diagram.
Persons who are G2a1a are confirmed as such by special SNP testing. Persons who are G will all be
positive for the M201 SNP. In addition, those positive for P287 are G2, and those also positive for P15 are G2a.
The P287 SNP is a rather recently identified SNP, and most of our samples were not tested for it. However,
anyone who is P15+ will necessarily be P287+. There are two further tests to bring the subcategorization
down to G2a1a. First P16+ indicates G2a1, and P18+ indicates G2a1a. However, the testing history of
P16 and P18 have been rather stormy and murky. It is my opinion that all the men on this page are G2a1a
though some have had results which showed P16+ but negative for P18. And one G2a1a man was found
negative for P18 at one lab, but positive for it at another lab. Another who seems to have similar marker values
to the men on this page was found negative for P16.
In 2007, researcher Thomas Krahn announced some problems with the SNPs themselves that may cause persons
who should be positive to lack the changes on the DNA that would be expected.
http://archiver.rootsweb.com/th/read/GENEALOGY-DNA/2007-07/1185576938
For a time Family Tree DNA removed P16 and P18 from its available tests. But apparently Krahn was not
able to persuade other labs to follow suit, and the tests were restored. A SNP should have as a
characteristic an inability to be lost, but Krahn's research indicates P16 and P18 are located on a part
of the chromosome where changes can occur. Despite the deficiencies of P16 and P18 testing, those men
who are found positive for them provide valuable information.
The reason the men here are all considered G2a1a even if -- in a few cases -- they were found negative for
P18 relates to the fact that logically they should be P18+. They are close enough genetically to proven P18+ persons
based on marker values that the SNP results should show P18+. With more samples, it can be better determined
whether allthese men are G2a1a (P18+) or whether instead there are men who are just P16+.
The smaller samples below were apparently gathered by Nasidze and colleagues in two studies of the Caucasus. There was
likely an attempt to sample men from more traditional communities and avoid recent migrants. He found most of the
men in the sampled communities in North Ossetia, Russia, belong to haplogroup G and have the highest percentage of G
in the world. Most of the G samples from these Ossetian towns are presumably G2a1a. There is also a striking
uniformity to the value combinations, which suggests a Middle Ages common ancestor for many of them, as is seen only
among the haplo G Ashkenazi Jews who suffered severe population losses in the Middle Ages and among the Welsh-Swiss
groups who share common ancestors in the Middle Ages. There is also a Georgia sample listed below. It is unclear where
this was obtained. So it could have been obtained in South Ossetia, which is technically part of
Georgia, but much of the population there refuses to be part of Georgia. We know also that pockets of G persons
were found in the eastern end of the Caucasus in a recent study, but it is presently unknown if any are G2a1a. If the
latter mentioned G persons turn out not to have G2a1a men among them, then this page here should be more logically
titled Ossetian men rather than men from the Caucasus Mountains. The marker value combinations on this page are
extremely rarely seen outside the Caucasus Mountains. Those G2a1a men found elsewhere seem to share a common
ancestor with the Caucasus men at least 2000 years ago, sometimes longer. While the people of North Ossetia refer to
themselves as descendants of the ancient Alans, the relative rarity of G2a1a outside the Caucasus suggests the large number
of Alans who invaded most of Europe during the Roman Empire had only a small contingent of G2a1a men with them
-- presuming there were G2a1a men among these invaders. There is a region in Hungary settled by Alans in the
Middle Ages. No samples from there are currently available to allow comparison with the Caucasus samples as it is
still not clear where these men resided before settling in Hungary.
The samples from Ossetia below from the YHRD database apparently have their origins in Nasidze's 2003
study "Testing Hypotheses of Language Replacement in the Caucasus: Evidence from the Y Chromosome," Human
Genetics, 2003, vol. 112, pp. 255-61. The data from his 2004 study are apparently unpublished. "Genetic Evidence
concerning the Origins of South and North Ossetians," Annals of Human Genetics, 2004, vol. 68, pp. 588-99.
This latter publication's data are also not in YHRD. I have seen the data from the Ossetian towns of Alagir, Zamankul
and Ziga. Almost all the samples are obvious G2a1a ones with marker values similar to those in the other Ossetian
towns shown below.
Some marker values from research studies are shown below in italics. These markers were not tested, but
the italizicized presumed values entered are either highly predictable or found in all G persons. The YHRD database
is a public database of samples collected from around the world as submitted by scientists. Marker values
in italics are only predicted and represent very predictable values.
The Stougarov and Parker samples technically do not have Caucasian origins, but the close genetic distances to those
who do makes it logical to display these samples here.
For information on what is known of the origins of this grouping here and migrations and time
relationships to other G groupings link to the migration-time page .