GENEALOGY-DNA-L Archives

Archiver > GENEALOGY-DNA > 2006-05 > 1147964352


From:
Subject: Rethinking Non-Surname Matches
Date: Thu, 18 May 2006 10:59:12 EDT


I’d like to suggest that it’s time rethink our approach of dismissing
matches with other surnames. I’m suggesting this for 2 reasons. (1) the extension
of testing to 37 and 59 markers brings the MRCA time horizon much closer and
consistent with the use of surnames (last 500 years) and (2) the rate of
non-paternity events is much higher and much more probable than first realized.

For example, if one reads the FTDNA position on non-surname matches,
_http://www.familytreedna.com/trs_explain.html_
(http://www.familytreedna.com/trs_explain.html) the basic argument is that there is too little data (too
few markers) to dismiss paper genealogies. For example, in the extreme case
of 1 marker, huge segments of the population match and DNA testing doesn’t
really tell us anything.

Specifically, the FTDNA site states:

“If you compare the 12 marker result to someone else who does not have the
same surname, but the scores match, you are most likely NOT recently related.
When we use the term recently related, we are talking about a time frame
within the last 1000 years or 40 generations, a time depth that accommodates the
earliest known use of surnames.”

It also should be noted that this page was written when 12 marker testing
was common and 25 markers was state-of the-art. A 12/12 match had a 90%
probability of a MRCA of 24 generations (840 years) which is at the cusp of when
surnames were becoming popular.

The page goes on to state that for a small number of markers, “the
importance of a surname link is paramount to provide a comfortable conclusion of
relatedness. Most of the time random matches with people with different surnames
do not stand the test for extended DNA testing.”

However, as the number of markers increases, this argument no longer holds
water. As the number of markers increases, the fidelity of the analysis will
approach CODIS and forensic analysis and high resolution non-surname matches
will become as relevant as surname matches are today.

For example, the table below shows the MRCA of perfect matches at a 90%
confidence level. Clearly 12 markers can do no better that take us to the time
when surnames were being created, and 25 markers don’t even bring us to the time
when most people’s paper trails hit a brick wall. (From my experience, most
paper trails hit brick walls in the 1700s.)

90% likelihood of MRCA
12/12 24 generations 840 years “almost beyond surnames”
25/25 10 generations 350 years “beyond most brick walls”
37/37 5 generations 175 years “within paper trails”
59/59 ? generations <100 years “within paper trails”

I think that most people would agree that in the future when 100 and 150
marker tests are common, we will need to look closely at all matches, not only
those within a surname. However, in looking at this table, it should be noted
that 37 and 59 marker tests bring us well within the timeframe of paper
genealogy.

So, when high resolution 37 and 59 marker tests conflict with paper
documentation – what gives? I believe that one aspect of matching that we tend to
ignore is non-paternity events. The occurrence rate of NPE’s is much greater
than generally acknowledged and frequently ignored by genetic genealogical
analysts.

My proposition is that not only as the number of markers increases, the
importance of matches across surnames becomes much more relevant, but that 37 and
59 marker testing brings us to this point today.

Thoughts?


This thread: