Group 2

DNA Analysis

DNA Results

Group 2a

All the members of Group 2a cluster fairly tightly in terms of genetic distance, which is never greater than 6 between any two members on 37 or 67 markers. All are within a genetic distance of 4 from the 2a modal 67-marker haplotype.

Group 2b

The average genetic distance between any two individuals in group 2b is 3 or 4, which is consistent with a common descent for the whole group within a genealogical timeframe. Member 97806 is a genetic distance of 6 to 8 with the other members. This is unusual, given that he shares a common ancestor with SOR3 in only 3 generations; he has had an abnormally large number of mutations in the recent generations of his lineage on the markers tested by FTDNA. Six of the seven members of this group (including 97806) have a documented line of descent from Samuel Vance/Agnes Penquite within about 8 generations and this is not contradicted by the DNA evidence.

Relationship between Group 2a and Group 2b

Although the genetic distance between individuals in groups 2a and 2b ranges as high as 10 on 67 markers, there is only a genetic distance of 4 between the modal haplotypes of the two groups, consisting of the following marker mismatches:

This was the genetic distance separating the two groups’ ancestors in the 18^th century. For instance, it was the genetic distance separating Abner Vance of Group 2a and Samuel Vance of Group 2b (we are able to deduce the ancestral haplotypes of both men). This genetic distance is small enough for us to conclude that Group 2b and Group 2a probably share a common descent from someone of the same surname or a variant. Two of the markers on which they differ (576 and 570) are among the fastest-mutating markers available for testing, while another (449) is also classified as relatively fast. However, this much genetic distance usually represents an accumulation of mutations over several generations, meaning that, however the groups split apart, the split probably occurred several generations prior to the earliest known ancestors in each group.

The following chart shows the genetic distances between members of Group 2a and 2b:

There are two broad possibilities for how Groups 2a and 2b might be related to each other: either they are on distinct branches or one of them is a subset of the other. In other words, the genetic distance that separates their modal haplotypes either accumulated in both lines after an initial split, or one of the lineages split off from the other and proceeded to accumulate mutations along the way to its MRCA. These scenarios are depicted on the following charts:

Of these scenarios, the first two seem more plausible than the third, only because Group 2a contains a much wider variety of lineages and therefore appears to be older. However, it is also possible that 2a and 2b are equally old and that 2a simply had more surviving lineages.

The closest match for Group 2 within the Vance DNA Project is Group 1, which is a genetic distance of roughly 16 to 18 on 67 markers from group 2a and a genetic distance of 13 on 67 markers from Group 2b. This indicates a low probability of a common descent within a genealogically meaningful time frame of the last 500 to 700 years, which is how long the Vance surname is thought to have been in use. It is only marginally possible that Group 1 and Group 2 share a common ancestor within the larger British de Vaux timeframe of 1000 years.

Matches with Other Surnames:

The 12-marker modal haplotype of Group 2 is identical to the Western Atlantic Modal Haplotype (WAMH), which is the most common Y-DNA signature among males of western European descent. According to FTDNA, roughly 2.5% of western European males have this precise 12-marker haplotype. Group 1 also has the WAMH; it is only possible to reliably distinguish these three Vance groups from one another on 37 markers or more.

Being WAMH, Group 2 has hundreds of close matches on 12 and even 25 markers with men in FTDNA’s database who have other surnames. On 37 markers the group becomes more distinct, largely due to their extremely rare DYS456=19/20.

When searching for those within a genetic distance of 6 on 37 markers in Ysearch, two men named Owens and one man named Whalen turn up as a match to members of Group 2a. These men also have certain diagnostic markers in common with Group 2a, which indicate a high likelihood of a recent common descent. These men have allowed us to incorporate their results in the Vance DNA Project, but I have listed them separately from the Vance members and lineages above.

• W1 – Patrick Whealen (b.1816, Ireland)

  • Member W1 has the surname Whalen. He is able to trace his descent back to a Patrick Whealen, who was born in Ireland in 1816, possibly in Tipperary, and who immigrated to Ontario, where he died in 1874. As yet, his DNA does not match any other Whalen men.
• 140425 – unknown (prob. a descendant of Matthew of Pittsylvania)
  • Member 140425 is an adoptee and does not know the identity of his father. He has the telltale 456=20, which is so far known to be held only by the descendants of Matthew Vance of Greenville and Robert Vance of Carter.

Observations on Marker Values:

The Group 2 modal haplotype is distinguished by one extremely rare value on the first 37 markers:

456=19 is extremely rare, and is highly predictive of membership in Vance Group 2. In group 2a, the descendants of Matthew Vance of Greenville Co, SC, and Robert Vance of Carter Co, TN, carry the even rarer DYS456=20. This value was not found at all in Leo Little’s survey of marker values. Its rarity is confirmed by a search of the database of the Sorensen Molecular Genealogy Foundation, in which only 4 men have this value, constituting a mere 0.014% of the men tested on marker 456. Three of these four men are actually descendants of Robert Vance of Carter County, TN, which makes the percentage of distinct occurrences of this mutation even rarer. It is so rare as to be virtually unique.

Group 2a and Group 2b each also have another rare marker on the first 37 markers:

As noted above, these two results serve to distinguish Group 2b from 2a. Group 2a and 2b share several unique values on markers from the final panel of 30 markers offered by FTNDA:

The sharing of these six rare values is one of the primary reasons the match between groups 2a and 2b is meaningful – it represents the coincidence of half-a-dozen rare occurrences, in addition to the rare occurrence of having the same surname.

Possible Shared Mutations:

456

The descendants of Robert Vance of Carter County, NC, and Matthew Vance of Greenville Co, SC, share the value 20 at DYS456. The traditional evidence suggests that Robert was the son of John Vance of Burke County and that John and Matthew were the sons of Matthew Vance of Spartanburg/Pittsylvania. The shared value of 20 represents a mutation away from the ancestral value of 19, and that mutation must have occurred either with Matthew of Pittyslvania or with some ancestor of his who falls in between him and the MRCA of Group 2a. Given the extreme rarity of 20 at DYS456, anyone who has this value stands a good chance of being a descendant of Matthew Vance of Pittsylvania or one of his close antecedents.

The value 18 at DYS456 may be a shared mutation between N39890 and 82984. The 18 held by 48374 at this marker is likely a parallel mutation, considering that he is 392=12, while the others who are 392=12 have 19 rather than 18 at 456.

Caution must be used when interpreting results for DYS456 in this group, as there is evidence to suggest that high allele repeat counts result in more frequent back mutations. We can have more confidence in shared values at the limit of the marker’s range, like 20 on 456, but when it comes to lower alternate values (such as the 18 shared by N39890 and 82984), it is more difficult to assess the likelihood that they are shared mutations.

The scientific literature contains several references to the phonemenon of marker instability when an allele repeat number becomes very high:

Dieringer and Schlotterer, 2003: “. . . the DNA replication slippage rate seems to be dependent on the length of the microsatellite. Alleles with a high repeat number are less stable than those with a small repeat number.

Lai and Sun: “When slippage mutations happen . . . contractions occur more frequently if the number of repeat units is large. When mutations happen, long microsatellites are likely to mutate to shorter ones; short microsatellites are likely to mutate to longer ones. The scarcity of large number of repeat units in a microsatellite locus can be explained by the high mutation rate and downward mutation bias when the number of repeat units is large.”

We may be witnessing this phenomenon at work in Group 2a on marker DYS456. Based on the conclusions arrived at in these studies we might expect to see an increased number of downward mutations in the lineages descended from the individual who originally had the mutation to the abnormally high value. That is just what we seem to be seeing in the Matthew of Pittsylvania group, where we have one confirmed back mutation to 19 in the lineage of Samuel Vance of Buchanan County, and two potential ones in the cases of Abner Vance and David Vance. The practical implication of this for an analysis of Group 2a is that marker DYS456 can not really be used to discount a relationship.

439

Members 82984 and Anc2 share the value 13 at DYS439. This represents a shared mutation that must have occurred with their common ancestor, Richard Vance of Union County, OH, or with some ancestor who falls in between Richard and the MRCA of Group 2a.

392

The value 12 at 392 may be a shared mutation between 48374 and 72197. DYS392 is a relatively slow-mutating marker, so the likelihood that the mutation is shared is probably greater than the likelihood that there were independent mutations to that value, though both scenarios are possible. W1, who is descended from a man with the surname Whealan, also has 392=12.

Haplogroup:

Both W1 and 56954 are positive for the L21 SNP, which defines a subclade of the P312/S116 branch of R1b.

The International Society of Genetic Genealogy maintains an up-to-date SNP tree on the R haplogroup here: http://www.isogg.org/tree/ISOGG_HapgrpR09.html.

Other resources:

R-P312 and Subclades Project: http://www.familytreedna.com/public/atlantic-r1b1c/default.aspx

R-L21 Project: http://www.familytreedna.com/public/R-L21/default.aspx?section=yresults