Archiver > GEN-MEDIEVAL > 1996-02 > 0823481173

From: Al Williams <>
Subject: Genetics and Genealogy (was Re: Adultery)
Date: Sun, 4 Feb 1996 17:46:13 -0700

>First, if a chromosome tends to be transmitted whole,
>then we are totally unrelated to at least a quarter of
>our sixth generation ancestors, receiving some 46 chromosomes
>from up to 46 of our sixth generation ancestors, but receiving
>none from at least 18 of them. The percentage drops rapidly
>with increasing generational distance because the number of
>chromosomes is fixed.

I think it is generally recognized that a chromosome is not transmitted
whole. A "centimorgan" is defined as "the unit of genetic distance, a
measure of how frequently two genes on the same chromosome are separated by
crossing over and therefore not inherited together. A distance of 1
centimorgan between two genes means that they have a 1 percent chance of
not being inherited together. In humans 1 centimorgan corresponds roughly
to a physical distance of 1 million base pairs."

>Third, if chromosomes tend not to be transmitted whome,
>the same argument holds with respect to genes or alleles,
>except that the number of generations to the point of
>"vanishing bloodlines" is a little larger: twenty generations exhaust
>a million genes.

I guess it depends what you are looking for. If you are looking at the
entire genome, as you could to establish paternity, you can go much further
back than if you are concerned with genes and inheritable traits.

I am not a geneticist, but have been interested in the relationship between
genetics and genealogy ever since I calculated that I had billions of
ancestors from Charlemagne's generation but only about a hundred thousand
or so genes to have inherited from them! I understand that there was a
significant amount of intermarriage in the past (my database is full of
families where I am descended from all the children of a medieval family),
and that there weren't that many people back then. But it has convinced
me, genetically, to think of myself as emerging from "the gene pool" rather
than specific ancient individuals.

This past year I have had the good fortune to work on the Human Genome
Project. For those of you not familiar with it, we are working to sequence
the entire human genetic code. I'm trying to use this opportunity to learn
about genetic inheritance from my colleagues. I'm still just beginning,
but let me share some of what (I think) I have learned so far.

* The human genome consists of about 3,000,000,000 base pairs (bp).

* 90% of these are contained in non-coding repeating sequences or "introns"
(often referred to as "junk DNA").

* Only about 5% of our base pairs (or nucleotides) are involved in protein
coding ("exons").

* These are the nucleotides that form the approximately 100,000 genes that
build us. (Genes are from 2,000 to 2,000,000 bp in length, averaging about
1,200 bp. or so.)

* Only about 20% of our genes have more than one possible allele (or
variant). Simple examples are the alleles for eye and hair color,
baldness, and sickle cell anemia. Most traits, such as height, are the
result of a combination a genes. The rest of our genes are the same in
each individual and are the "housekeeping" genes that must be present and
that cause us to have a head, a liver, two legs, etc. (This is why you
read that the genetic difference between humans and chimpanzees is only
about 1.2%. They also have a head, liver, legs, etc.)

* I haven't mentioned base pair substitutions or mutations. This occurs
about once for every 600 base pairs. I'm not sure how to factor this in
yet. Those that don't kill you will likely become extinct in a few
generations anyway.)

So, now we're talking about approximately 20,000 variables of inheritance.
That's still a lot of different combinations! I don't know how many
different alleles (variants) there are at any particular site - 2, 3, 1000?
And I don't know how some of the non-coding nucleotides might still factor

If we have 32,768 ancestors, not counting intermarriages, just fifteen
generations back, the odds of finding phenotypical similarities that might
have come from one in particular are not good.

Again, the whole 3 billion base pair sequence can be used to prove
ancestry. That would allow us to go a little further back. But if "Junk
DNA" is just that (or even if it's not), beyond a certain point whether or
not Exon the IX was the son of Exon the VIII or the milkman may not really
matter that much now.

Let me say again that I probably have some of this wrong. The science is
still young and I suspect that all these figures are only a little better
than educated guesses. I'm pretty sure I can get different figures for
each of the points above depending on where I look. If you see anything
grossly incorrect or can add anything to what I have written please write
to me or the list.


Al Williams

This thread: