Re: [asa] Two questions... (bottlenecking)

If you go back far enough, you should have a single ancestral pair for
all modern humans. However, "far enough" might involve millions of
years, considerably more fur and less brain, living in trees, having
tails, etc. On the other hand, it's conceivable that a scenario like
Glenn Morton's, with Adam and Eve ancestral to all modern humans about
5 million years ago, would work genetically.

On p. 126 of The Language of God, Collins cites population genetics as
evidence that modern humans descend from a group of about 10,000
people about 100 to 150kyr ago. RTB favors a separate creation of
modern humans from Neanderthals, etc., and would want a population
size of about 2 at the same time period. What I'm saying is that you
probably had more than a single pair at that point, but might
conceivably be able to get to a single pair at some earlier time, if
you are comfortable with pushing Adam and Eve back that far.

Collins knows a lot more than me about human genetics; I have more
academic training in evolution than he does. As a paleontologist who
does DNA work because it a) answers questions I'm interested in and b)
seems to provide slightly more prospect of eventually getting a
regular income, I tend to be rather more skeptical of some of the
claims based solely on DNA evidence than the average person with a
strictly molecular biology background. Take those into consideration
in evaluating the comments.

Assumptions that go into the estimate of 10k people at 100k years:

Proportion of males and of females who contribute to the next
generation. If only a few individuals (at least of one sex) are the
parents of most of the kids, then the effective population size will
be a lot smaller.

Average genetic diversity of the starting population.

Any unusual events, such as a population crash and rapid rebound.

Mutation rate. Calibration of this (i.e., assigning dates to a
particular level of divergence) is often very poor. After all, what
you need is a very precise date for when two individuals last shared a
common ancestor.

Selection pressure on the changes under consideration. For example,
the point of the MHC locus genes is to keep changing to stay one jump
ahead of the pathogens. This favors lots of rapid change. The 18S
gene is an essential part of making proteins. Many parts of it show
very little change. Other parts of the genome are apparently fairly
free to change without having much effect, but there may be structural
reasons why some of those regions change more than others.

Change the details of the assumptions, and you can get different
results. In fact, it often seems as though you can get just about any
possible result if you try enough population genetics models.

One last complication, related to the mitochondrial Eve thing, etc.,
is that you don't have to go back too far in time (rather less than
100 kyear) to have one pair who is _an_ ancestor of all modern humans.

I think Adam and Eve as representatives out of an existing human
population is the easiest way to reconcile Genesis and genetics, but
it is not absolutely impossible to have a single pair ancestral to all
humans if you go far enough back in time.

-- 
Dr. David Campbell
425 Scientific Collections
University of Alabama
"I think of my happy condition, surrounded by acres of clams"
To unsubscribe, send a message to majordomo@calvin.edu with
"unsubscribe asa" (no quotes) as the body of the message.