J2 is a Mediterranean /Neolithic Haplogroup, thought to have been dispersed with the spread of farmers from Anatolia during the Neolithic agricultural revolution. Broadly speaking, the greatest percentages of Haplogroup J2 are found in populations in Anatolia, the Caucasus, and in coastal areas around the North and North Eastern shores of the Mediterranean sea (eg. Southern Italy, Greece ). It is J2's ?brother haplogroup? J1 that is most common in the fertile crescent.

Much new knowledge has been gained in the nearly two years since I first tested and learned that I am Y-DNA haplogroup J2. For instance, we now know that Haplogroup J2 is further divided up into several
sub-clades, and recent studies have started measuring the percentages of the separate sub-clades in different geographical areas. Recently I have learned that I am haplogroup J2f* (now known as J2a1b*), which Semino et al. (2004) speculates might have been dispersed through the Mediterranean by seafaring neolithic populations. In future we hope that as knowledge expands on the geographical distribution of the separate sub-clades, that we will gain an even better understanding of the population histories of our paternal line ancestors.

The project's aims are to:

• Gain a better understanding of how all Haplogroup J2's are related to other Haplogroup J2's

• Expand the knowledge of what haplotypes are found in the various different sub-clades of J2.

• Create a database of results that is guaranteed to be more accurate* than those in user-submitted

public databases.

• Determine how accurately sub-clades can be predicted from haplotype markers, using scientific methodology.

The Haplogroup J2 "family tree"

Haplogroup J2 is one of the several branches on the global Y-DNA family tree. Haplogroup J2 is defined by the SNP marker M172, and all men that have the M172 marker are descended from a common ancestor who is thought to have lived about 15-22,000 years ago in either the Levant (Semino et al. 2004) or Anatolia (depending on how the J2 diversity reported in Di Giacomo et al. (2004) is interpreted).



Many haplogroups - including haplogroup J2 - can be split into sub-clades. At various different times in the past, a M172+ man has gained an additional DNA change (SNP's) that will be shared by all his descendants. Some of those extra changes have been found by geneticists, and help us divide J2 into various different branches of the overall J2 family tree. In future more of these SNP's will be discovered, which will enhance and expand our understanding of Y-DNA haplogroup J2.

How can I tell what subclade I am ?

The STR marker tests (eg. 12, 25, 37 markers) used for genealogy give you your haplotype. From haplotypes quite often your haplogroup can be predicted reliably. Similarly, in some haplogroups the haplogroup sub-clade can also be reliably predicted from your haplotype. However, haplogroup J2 is a haplogroup in which we have not been able to reliably predict the sub clade from haplotypes (at least, not from the short haplotypes that population geneticists use). Haplotypes may be similar/cluster together, which implies they are closely related, however sometimes the true biological relationships do not correlate with overall haplotype similarity.

The reason why the haplotype sometimes doesn't correlate with the true relationships between lineages is because of what is known as haplotype convergence. This is because of the different ways the two different types of marker change.
Basically the haplotype "distribution" within one lineage overlaps with the haplotype "distribution" of another lineage - its like overlapping branches from two different trees.

The fewer the number of markers in the haplotypes we are comparing, the higher the chance that we will see convergence. Conversely, the greater the number of markers in the haplotypes we are comparing, the lower the chance of convergence happening.

What does the project hope to do?

In haplogroups such as I, the sub-clades can usually be distinguished reliably from haplotypes, and usually analysing the haplotypes will result in identifying clusters that do more-or-less represent true biologically related lineages. However, because of the degree of convergence that we see in 10 marker haplotypes in haplogroup J2 (and J as a whole), we cannot assume that analysing haplotypes in the absence of SNP data will identify true clusters of lineages. The only way we can be sure that clusters of similar haplotypes do represent true lineages is to compare the actual SNP results, with clusters identified from extended haplotypes (clusters identified by analysing with programs such as Fluxus, or by visual inspection of haplotypes).

This is the only way we can accurately determine whether any markers and/ or combinations of markers in the extended haplotypes do correlate with the true biological lineages. We need to be sure that any haplotype patterns we do see are real patterns rather than "random noise".

The project is now entering its second phase of haplotype analysis. In this second phase, allele frequencies will be analysed statistically, and aims to determine which markers can be used to predict sub-clades. In the third phase of analysis, the project will quantify how accurate sub-clade predictions are (by comparing the predictions with the actual SNP tested results) - ie. so we can say what percentage of the time a haplotype predicted as J2e (J2b in Sengupta et al. nomenclature) actually is a J2e haplotype.

When these results are conclusive (ie. can be regarded as fact rather than speculation and/or as yet unproven hypotheses), they will be publically presented. Scientific hypotheses and speculation are good and ultimately advance science, but sometimes when this is in the public arena it is mistaken as fact, and confusion can arise. The project doesn't want to suppress information, it just wants to make sure that all information publicly presented is fact. The project stands for the discovery and presentation of accurate information, not the perpetuation of misinformation.