A Scientific Perspective
January 15, 2007
People
have asked Michael why they should bother coming up
with $100-150 for a DNA analysis. He asked me to
summarize what the Ragsdale surname project has
found so far. This is my own take, so the
administrators of the project are free to disagree.
Originally, it was often said that every Ragsdale in
America descended from Godfrey Ragsdale who died in
the 1644 massacre, and from his only son Godfrey.
For most of us these men are some 10 generations
back. These are southern lines, whose records
suffered terribly during the destruction of the War
between the States as well as regular courthouse
fires. Thus many of us cannot prove our descent from
the Godfrey Ragsdales. The lines tend to peter out
in the early 1800s in any number of southern states.
That is where DNA comes in, because it provides a
nearly unvarying fingerprint of your Ragsdale
ancestors.
The simplest and cheapest DNA analysis costs about
$100 and gets you 12 DNA markers. These 12 pieces of
DNA were chosen because they are particularly
characteristic of the various genetic groups around
the world. If all the Ragsdales came from the
Godfreys, all the descendants should have nearly the
same numbers. (The meaning of the numbers is
unimportant genetic technology--we just want to see
if they are the same.) I say "nearly" because
changes, or mutations, can occur rarely but
naturally. They happen once every 500 generations or
so. Because we look at 12 markers, it is likely to
have a mutation in one of these 12 in 500/12
generations (about 40). Because Godfrey is about 10
generations removed from us, we are still well under
these figure of 40. Thus most descendants should
have identical DNA, but maybe a few will have at
most one mutation. (If you look at 25 markers, it
becomes 500/25 = 20 generations, so more people can
show one or two mutations.)
Right now seven men have had their DNA analyzed, and
we have four different patterns (the geneticists
then say we have four haplotypes). We have three men
with the same pattern (participants 54757, 56964,
and 72061), and one (48658) who differs at only one
marker from these three. Does that single difference
indicate his DNA experienced the one permitted
mutation? Look at the 25 marker results for these
men, and indeed the next 13 are all the same. If
48658 had been different, he would have shown many
differences in these markers. We can confidently say
that these four men are closely related, as they
agree at 11 of the 12 markers and at 24 of the 25
markers.
Participants 49632 and 57688 have the same set of
markers (12/12 and 25/25), and so are closely
related, but they differ at three markers from the
three men with the same pattern in the first group.
An agreement of 9/12 (and 18/25 if you look at all
25 markers) indicates little relationship. Their
common ancestor was several thousand years ago.
Participant 60715 has similar large differences from
all the rest.
I know nothing about the paper genealogies of these
men, but they cannot all descend from the Godfrey
Ragsdale. What then can we conclude? (1) More than
one unrelated Ragsdale came to America, and these
patriarchs produced the four Ragsdale haplotypes.
(2) There was what is delicately called a
nonpaternal event. That means that at some time a
male child purported to be on the Godfrey Ragsdale
line was not in fact sired by a Ragsdale. There are
three possibilities causes of an such event:
adoption, a name change, and illegitimacy. The
adopted male child would have the name Ragsdale but
the genes of his birth father. The same thing would
hold if a Ragsdale wife gave birth to a son by a
non-Ragsdale father. The name change usually occurs
during adulthood and would probably be recorded,
unless court records were lost.
For the Ragsdale project to work, we need the
participation of lots of Ragsdale men. If many have
good paper trails back to Godfrey, we can finally
say exactly which haplotype (marker pattern) was
Godfrey's. Right now I think it is the one shared by
54757/56964/72061. People who match that pattern (or
have an 11/12 or 23/25 match) know they descend from
Godfrey, even if they are unable to get around their
genealogical brick wall. For those who have
different patterns, we may identify other, unrelated
Ragsdale patriarchs (immigrants), or we may identify
exactly which lines had the nonpaternal events. The
bottom line is, the more participants, the faster we
will approach that goal. Also, there are Ragsdales
in England, and we need their participation too. One
or more of the non-Godfrey patterns may match up
perfectly with a living English Ragsdale, and those
men then will know where their line came from.
March 29, 2007
OK, here goes. We
now know for sure the haplotype of Godfrey Ragsdale,
even though we have seven different haplotypes out
of eight analyses. We have to go with the largest
group, and that contains five people. Your group,
49632 and 57688, has four differences in the first
12 markers from the Group of Five, and your group
and the Group of Five are both four different from
60715. For haplogroup R1b1, these are huge
differences, indicating very distant common
ancestors. The fact that you and 57688 have a 37
marker match means that there is a definite second
Ragsdale line, but not from Godfrey.
We can easily see the Godfrey haplotype by looking
at the first five lines in the results table.
Godfrey's DYS numbers are either five out of five
(the same for all members of the group) or four out
of five (differing by a single mutation for only one
member). Traditionally, the progenitor's haplotype
is called H1. We have three other haplotypes on
this line, which I am calling H2, H3, and H4, just
in the order they were analyzed. 49658 (H2) had a
mutation at 391; 56964 (H3) had two mutations, at
456 and 576; and the new 82433 (H4) had three
mutations, at 438, 570, and CDYb. Most of these
mutations are in the faster moving markers found in
the 26-37 test, so two or three such mutations are
not unusual for a line from the 1600s. Those
percentages you quoted assumed that every marker
mutates with an equal rate, but this is not the case
at all. To take into account the exact rate of
mutation for each one would be the accurate thing to
do, but impractical.
I
have drawn a chart which shows Godfrey at the top as
H1 and the other three haplotypes as divergent
lines. None of these other three haplotypes can
come from each other, as they have no common
mutations. I plan to update it as other analyses
come through
So, the project is going very well.
Joseph
Joseph
B. Lambert
Department of Chemistry
2145 Sheridan Road
Northwestern University
Evanston, IL 60208-3113
Telephone: 1-847-491-5437
Fax: 1-847-491-7713
http://www.chem.northwestern.edu/~lambert/
Many,
many thanks to Dr. Lambert for going above and
beyond to explain things ---- Michael
