|On the face of it, a Y-DNA STR haplotype appears to be just a string
of meaningless numbers — and without a lineage, it is. For a haplotype
to be useful, we need to understand what the values for the markers mean
when used in combination with paper genealogy. To simplify understanding
them, I place these marker values into four categories:
1. Modal Values. These values are the most
common ones for the entire haplogroup subclade; and, because they are the
most common values to have, possessing them is literally unremarkable.
To demonstrate the above, I've taken an example from a cohesive lineage
group in the GIBSON
project as shown in the table below (thank you, Valentine Van Zee,
Gibson Project Administrator, for your assistance and cooperation).
We are all related to one degree or another, so it would be superfluous
for me to say all the individuals in the table below are "related."
The question is whether or not they are related in genealogical time,
that is, since the adoption of surnames and the keeping of records for
common citizens. For most of us, that means being related within
15 to 20 generations or within about 450 to 600 years.
2. Signature Markers. These are non-modal marker
values shared by everyone in a family. They consistently and in
combination uniquely distinguish a family. Presumably, these
are the mutations possessed by the family's progenitor within "genealogical
time." In contrast, sharing the haplogroup modal values with a family
is of virtually no significance beyond the fact that you're related somewhere
back in time — as we all are if you go back far enough.
3. Private Markers. These markers are unique to one
test subject. The mutation has happened in the individual's descent
from the family progenitor, but we don't know in which generation.
As more cousins are tested, private markers will turn into branch markers,
unless the mutation occurred in the test individual, himself. Because
the latter can be the case, it is not irrelevant to test a brother, father,
or first cousin. Anyone with a private mutation should test cousins
(viz., a 1st cousin, 2nd cousin, 3rd cousin, etc.) until the location
of the mutation is determined.
4. Branch Markers. These are marker values shared
by two or more members of a family, but not the entire family, in other
words, by a branch of the family. Once someone has proven their membership
in the family by possessing its signature markers, the branch markers become
highly important because these markers most likely represent a shared common
ancestor more recent than the family's progenitor. (In some cases,
a shared value on a volatile, fast mutating marker may mean the mutation
happened more than once in the family; paper genealogy and the testing
of cousins can determine which is the case.)
|One thing obvious from the table below is that everyone needs to upgrade
to 67 markers to really understand what is going on. The GIBSONs
who have not tested 67 markers will have to remain in an "unassigned" group
until they upgrade. I have found the need for 67 markers to be the
case for Haplogroup R1b in all my projects. If a haplotype is rare,
you can sometimes get away with fewer markers, but not in R1b. Haplogroup
R1b is the most common haplogroup in western Europe, reaching a frequency
of 70-80% in the British Isles (please see distribution
map). Sixty-seven markers is minimal for an R1b.
The first row in the table below is the modal haplotype for Haplogroup
R1b, which is highlighted in cyan (this shade of blue). Like most
individuals, the test subjects match their haplogroup's modal haplotype
at most markers. Being R1b and matching a modal value for R1b should
be considered of minimal genealogical significance. It simply supports
that you are a member of a large group originating tens of thousands of
years ago. The genealogically significant values are those that differ
from the R1b modals, so please ignore all the values highlighted in cyan
and consider the sea of blue simply background noise. My use
of the R1b modal should not be taken as implying that it is the ancestral
haplotype for either the haplogroup or the family, nor that the genetic
distance from it can be used to estimate the distance in time from the
haplogroup's common ancestor. There are many paths from the haplogroup's
ancestor to its descendants, none of them direct (i.e., without
innumerable twists, turns, and reversals — on each and every marker).
With the aid of DNA test results and paper genealogy, we can, at least,
hope to unravel the last fifteeen to twenty generations.
Ignoring the markers highlighted in cyan, we are struck by the agreement
of most of the differences, that is, the six columns highlighted in the
muted orange color I've selected for the GIBSON family, plus two columns
highlighted in bright green. These are this GIBSON family's signature
Scattered around the table are cells highlighted in bright yellow.
These are private mutations, ones acquired by these individuals in their
descent from the family's progenitor. When testing 67 markers, you
can statistically expect one mutation event to occur about every
seven generations, and experience has shown (empirical evidence demonstrates)
that the number of mutation events typically ranges from 0 to 3 in an individual's
descent from a common ancestor
in genealogical time. (This
number can be higher in families with deep roots, such as some Scottish
clans with paper genealogies tracing back into the 13th Century.)
I've highlighted the branch markers in purple. The mutation at
DYS413a represents a major division in this GIBSON family. If everyone's
pedigree is carried back far enough, the lines will presumably converge
on the ancestor in whom this mutation occurred.
|There are three other cases where this GIBSON family has individuals
sharing what would otherwise be a "private" mutation (i.e., a value
that is neither modal for the haplogroup nor signature for the family).
In one case, it's clear that it's a true branch mutation, that is,
a case of two individuals sharing the same common ancestor, one more recent
than the family progenitor. In several cases I suggest this is a
"possible" branch marker because the mutations are on volatile markers.
We'll need confirmation from paper pedigrees to confirm whether the match
reflects relationship or a simple coincidence.
|• The certain branch marker is shared by G-6 and GX-1 who have
a value of 14 at DYS392 when everyone else is 13 at that marker.
A check of their pedigrees suggests John (G-6) is the great-grandfather
of Gideon (GX-1). It appears a value of 14 at DYS392 will identify
descendants of John, with the inference that no one else in the table descends
|• The value of 41 at CDYb for G-20 and G-23 does not appear to
be a branch marker. CDY is a volatile multi-copy marker; and, because
G-20 and G-23 differ at DYS413a, I believe this is a case of the
same mutation happening twice on CDYb, making each of these a private marker,
not a branch marker. Note that these two individuals cannot be placed
on a logically branching tree unless either the CDYb mutation happened
twice or the DYS413a mutation happened numerous times. As the former
is far more probable, I'm assuming that is the case here.
|• There are possible branch markers at DYS576 where a value of
19 is shared by G-106, G-113, and G-11, while a value of 17 is shared by
G-23 and G117, and at DYS391 for G-117 and G-106 who share a value of 10.
However, all three mutations cannot be branch markers and still be placed
in a logically branching tree. As DYS391 is a fairly stable marker
and DYS576 is rather volatile, it's more likely the DYS391 mutation is
a branching marker, while one or more of the DYS576 values are private.
The paper pedigree and testing cousins can resolve which.
|Panel 4, the 38-67 marker panel, is generally considered the most stable
of the four panels, so many consider it not worth testing because of its
lack of variability. While our GIBSONs follow the pattern in being
less variable in Panel 4, the fact is we find the single most important
branching marker in the family in Panel 4, namely, the values of 22 vs.
23 at DYS413a. Had most members not tested Panel 4, it's significance
for the family would likely have been overlooked.
|Two individuals, namely the HURST and the COLLINS, appear to have NPEs
in their lineages, that is, they are really GIBSONs. Ironically,
each of these matches 67/67 to the GIBSON modal haplotype (and, no, I don't
see any particular significance to that fact).
|It appears the JUSTICE individuals are related to these GIBSONs, though
it's unclear whether it's through an NPE
or simply that their connection is prior to surname adoption, though I
would bet on the latter. In a cladogram, their haplotype would be
the most ancestral one, which is the reason they are placed at the top
of the table. The ancestral haplotype is followed by the addition
of the mutations at DYS449 and CDYa (bright green table cells), then followed
by the mutation at DYS413a (dark purple table cells), then followed by
the "private" mutations (bright yellow table cells). If you consider
these GIBSONs and JUSTICEs to be just one family, then the markers highlighted
in green switch from being GIBSON signature markers to being GIBSON-JUSTICE
|Bottom line: everyone needs to upgrade to 67 markers!
And the person tested elsewhere (GX-1) needs to join the project at FTDNA,
then upgrade to 67 markers.