GENEALOGY-DNA-L Archives

Archiver > GENEALOGY-DNA > 2002-12 > 1040676960


From: "John F. Chandler" <>
Subject: Re: [DNA] Counting ancestors, measuring ancestry
Date: Mon, 23 Dec 2002 15:56 EST
In-Reply-To: jerrys3@earthlink.net message <5.1.0.14.2.20021222110558.0266bc80@mail.earthlink.net> of Mon, 23 Dec 2002 10:47:44 -0700


Jerry wrote:
> Some years back I recall a book (perhaps "Heredity and You"?) that said it
> was possible to *not* be genetically related to a grand-parent and that by
> the sixth generation back, it was around a 50/50 chance that there was a
> genetic connection with any one ancestor.

That would be true if each chromosome were an indivisible unit (and
neglecting the X and Y chromosomes). However, chromosomes are not
indivisible, so that calculation is worthless.

> It also said that most genetic material was passed along via whole
> chromosomes but there was some 'cross-over' (don't remember what that was)
> possible.

Cross-over is a process that causes the two chromosomes in a pair to
"break" into two pieces at the same spot and then "rejoin" to the
opposite one. This happens relatively often. This means that two
genes on the same chromosome can be separated in the descendants, and
the likelihood of separation increases with the distance between them
on the DNA chain.

> If so, might not such a 'loss' of ancestors account for at least some of
> the discrepancy between our family knowledge of heritage and that measured
> by DNA testing

If you're talking about the percentage ancestry analyses, then that's
exactly right. Even though the complete "loss" of an ancestor takes
longer than that simplified calculation would indicate. each ancestor
does dwindle away as the generations pass, and the cutoff occurs when
the contribution of any one ancestor is below the "measurement error"
of the analysis.

John Chandler

This thread: