PACE-L Archives

Archiver > PACE > 2002-09 > 1031407661


From: "Roy Johnson" <>
Subject: [PACE-L] DNA
Date: Sat, 7 Sep 2002 09:08:12 -0500
References: <3D76D573.824A5CCA@eagle.ca>


Gordon and I have joined the Genealogy-DNA-L mailing list, and I have
submitted many questions and have received some answers. I have also
browsed a few sites for those actually involved in DNA studies. Here are
some conclusions:

The bad news is that a DNA study can NOT help you to know, or even guess, at
who the immediate ancestors of your great-grandfather were. The tests
operate by looking for mutations in up to 25 Y-chromosome alleles, or
markers. The 25-marker test is the most expensive, averaging $250 per
individual. A perfect match on a 25 marker test of two individuals
indicates that they have a slightly better than 50% chance of being related
in 8 generations or less, ranging up to a 95% chance that they are related
in 40 generations or less. So regarding possible relationships of less than
8 generations---forget it!

A DNA study is most effective in looking for relationships between larger
groups, rather than individuals. Responses have indicated that a study of a
John of Middlesex-Richard of Jamestown descendents would be a perfect
example. In such a study, a less expensive 12 or 14 marker test could be
used (under $100 per individual), and might even be better than a 25 marker
test because the reduced expense would allow for more individuals to be
tested. A perfect match on a two-individual 14 marker test only shows
probability of a relationship some time since the Norman conquest (1066),
but if you have, say, 20 individuals from different branches of the John
line and 20 from different branches of the Richard line with perfect or
close to perfect matches, then this increases the odds that they are more
closely related.

The test is better at eliminating than confirming; for example, if the
results on a 12 or 14 marker test showed more than two or three consistent
mismatches between the lines, it is virtually certain that the two lines are
NOT related in any reasonably recent time frame.

If one individual gets a huge mismatch and the others being tested do not,
it indicates that he is descended from an unreported adoption or a hidden
illegitimacy--in other words, he is not a biological Pace. Some might not
want to find that out.

The 12/14 marker tests can be upgraded to a 25 marker test without
re-submitting samples, at a cost of about $180 for the upgrade.

These tests have medical benefits to the individual being tested. I have
not been able to explore this yet.

As to labs, the overwhelming vote (about 8 to 1) is for FamilyTreeDNA.com
(FTDNA). FTDNA uses the University of Arizona lab and gives a huge amount of
help in coordinating the project. There must be an administrator, and
participants must sign release forms allowing the administrator to see the
personal DNA data. FTDNA provides a secure password protected site for the
administrator to store information. The administrator assigns a number to
each participant, known only to the participant and the administrator. The
administrator prepares a chart of the results, showing the matches on each
marker, and identifying the participant only by number, so that each
participant can compare his results to the others but will not know who any
of the others are specifically. I found some examples of such studies but
forgot to bookmark them. I'll try to post them later.

I suggest those who are interested go to http://www.FamilyTreeDNA.com and
browse around. There is a neat tutorial on how such studies work, and there
are examples of other family studies.

In doing this research, I have neglected postings to PacePix. For those of
you waiting patiently since May, my apologies.

Roy Johnson

This thread: