GENEALOGY-DNA-L Archives

Archiver > GENEALOGY-DNA > 2001-12 > 1008179873


From: "Orin R. Wells" <>
Subject: Re: [DNA] BBC GENE STORIES ?
Date: Wed, 12 Dec 2001 09:57:53 -0800
In-Reply-To: <96.1ec76122.2948e2c0@aol.com>


I will confess on the front of my statement that I am not a geneticist, so
there is probably no scientific basis for what I am about to write.

At 11:41 AM 12/12/01 -0500, wrote:
>The Genetic Past Jim Wilson.
>"If a group of people all share a genetic change on part of their mtDNA or Y
>chromosome, we can assume that not only did they have a common ancestor<<

I personally do not fully agree with this. The assumption being made here
is that if you see a change on one marker that has mutated it necessarily
means that it ONLY happened to one individual in this family (assuming they
ARE only referring to a related family). I don't think you can really make
this assumption although the odds ARE in favor of it being a single occurance.

>>Focussing on just these parts of the DNA chain also allowed the age of
any changes in the genetic code to be estimated.<<

Again, it is ONLY an estimate. The ranges are often pretty wide -
thousands of years at times. It is like playing roulette. How often will
the number 13 come up on the wheel? If they wheel has spun that many times
without 13 coming up will it come up the next spin? Not likely. If it
were that simple we could all win the Lottery pretty easily. The problem
is that chance and random selection effects play into the equation enough
to make this far from being an exact science.

>>The number of accumulated changes in these types of DNA can be used as a
measure of time - the more changes there are, the further back in genetic
history we are
looking.<<

Again, they are making the assumption that if they see two changes in a
group that the changes occured an "average" number of generations apart.
They could just as well have occured in two generations in succession OR
even in the same individual. They really don't know what triggers such
changes. That individual could have lived the "estimated" number of
generations ago or he could have been your grandfather. Without a lot of
samples of the same group they are using their best informed guess. It can
be thrown off also by such things as this occuring in the youngest son of
the youngest son etc. or in the oldest son of the oldest son etc. which can
have a termendous impact on the average number of years represented by a
"generation". There are a lot of unknowns at work here, so to take these
estimates as fact is probably not valid. When they try to pin down "the
earliest ancestor" in Europe to be within a range of 20 to 30 thousand
years it tells us they have a long way to go before they can be precise.

>>We can discover just when the change we see happened, allowing us to
identify in which period of history the person in whom the change happened
lived."

I think that depends on the number of years they are allowing for "period".
They will have as much chance of winning the lottery as to be getting this
right.

>Does this mean that a person with, say, DYS 19, marker 14 has older
ancestors then a person with marker 13 ??<<

No. Without additional information you can not tell whether 14 was mutated
from 13 or 15 and the same with 13 (12 or 14). It could have happened with
your grandfather or 100 generations back. AND, unless you are looking at
a specific family group, there is no reason to believe the closeness of the
numbers in the absence of other matching information, are related at all in
recent history.

>How may one relate the above statements by Wilson with their own (DYS
numbers) Y-ch signature ? Also as to the mtDNA ?<<

I don't think you can. It is just interesting observations made by someone
in the infancy of the industry. Make no mistake, we are on the cutting
edge of this application of DNA. There is still a LOT to be learned.

All you can determine from your y-DNA is that you may or may not match
someone else. If you don't match, you are not closely related. If you
match on many markers, you are probably related somewhere in more recent
history but without supporting traditional genealogical documention you are
not likely to be able to figure out how. If you match exactly, barring any
mutations in recent history, you can say with pretty high probability that
you are closely related. But is it within 1 generation or 100? Again,
without supporting documentation you have no real way of knowing.

The mtDNA, since it changes so slowly, can tell you if you (and your mother
and your mother's mother etc.) are related to someone 40,000 years ago. I
am not sure what this does for one's family history search. But when you
realize that they essentially are playing with 7 to 20 ancient families
here, it means you will pretty closely match about 300 million other people
on the planet more or less. If you thought trying to remember the names of
the cousins you knew about was hard....

When they start getting into the Nuclear DNA (the markers carried on the
OTHER chromosomes) they will begin to give us ways to find relationships
that traditional genealogy does not always povide especially with lost
records or where none can be found. Each of these chromosomes is
contributed 50% by each parent. So, each child will have some mix of this
from each of the parents and usually not exactly the same mix. That is why
we all look different from each other. Each generation dilutes this by 50%
but we each carry markers passed on from our earliest ancestors. Once they
have done the research to identify the markers in this area that can be
used for genealogical research we should see some more precision in using
DNA testing to identify relationships other than direct Male or Female with
no allied branches allowed. That, I am afraid, is probably several years
down the road.


Orin R. Wells
Wells Family Research Association
P. O. Box 5427
Kent, Washington 98064-5427
<>
http://www.rootsweb.com/~wellsfam/wfrahome.html
Subscribe to the "Wells-L" list on RootsWeb

This thread: