Hello Peter,
You wrote:
>Is there any _direct_ evidence for common ancestry between humans
>and apes?
I recently bought a series of time-lapse photographs showing the
connection from a street vendor in New York how claimed they were
authentic. Does that count? ;^)
>Of course, there is a lot of evidence for this, particularly in the
>field of DNA sequences.
As well as the fossil record which provides information about when and
where the lines could have diverged. That information can be used to
calibrate and test the neutral mutation assumptions discussed
below.
> For various reasons, I am convinced humans
>and apes are biologically related by common ancestry, but I am not
>aware of any _direct_ evidence for this.
>
>Although the similarity between corresponding DNA sequences of, say,
>human and chimpanzee may be statistically highly significant, common
>natural selection may possibly account for this (parallel or convergent
>evolution). Therefore, the significance of the evidence for common
>ancestry is directly related to the _lack_ of functionality of the
>sequences under consideration.
I'll focus on one case.
The often degenerate, third base of codons in sequences encoding proteins
*may* have a selective effect in some cases. However, I think it would be
exceptionally unlikely that *all* (or even most) such positions were
subject to the level of selective pressure necessary to maintain such
a degree of similarity or drive such a level of convergence. I believe
that neutral theory provides a null hypothesis for sites thought to
have been subject to selection. Additionally, we know from experiments
in the lab and from sequences obtained from species with similar
habitats and morphologies that sequence _convergence_ down to the third
base in a codon is the exception, rather than the rule. On the other
hand, sequences linked by presumed common descent tend to exhibit greater
similarity at the third base.
>Now, a pseudogene or other unused sequence which has evolved for the
>last 6 million or more years in the ancestors of humans and chimps in an
>exactly neutral way is, presumably, under no selective pressure. But is
>it possible to demonstrate that such a sequence has not been used at any
>time in either line of descent for any function, coded or non-coded?
That goal is impossible to achieve even in studies of organisms of today.
As you know, we cannot subject an organism to all possible environments
and conclusively demonstrate that a change in a single base has absolutely
no effect on an organism's fitness. For example, in wild-type
S. typhimurium there had once been no documented examples of a growth
variation in glutamate dehydrogenase knockout mutants. I found one
growth condition unlikely to be encountered in the wild (alanine +
ammonium) where such a phenotype could be found. That was a single
gene knocked out with a few mutations. The amount of sequence divergence
within humans is on the order of about 0.2% (?If I recall correctly?).
We cannot evaluate the individual impact of those millions of base pairs
let alone the effects of their interactions.
[Aside: The discussion of within species variation is an important one
within medicine. For example, previously unsuspected and possibly,
previously unselected variations in enzyme activities can affect drug
clearance and toxicity. Now that we have finally developed some of
the primitive tools needed to addressing this medical issue, you
will see a lot more of this work coming out of the pharmaceutical
and biomedical research communities. It will be interesting to see
how much impact individually customized treatments could have on
medicine and agriculture.]
>Presumably, any single unused sequence will be too short to give a
>sufficient signal-to-noise ratio to provide a definitive answer.
>Therefore, large sets of such sequences would have to be tested
>together.
>
>Have any such "phylogeny-indicator" sequence sets been documented?
By the criteria you have chosen, no "phylogeny-indicator" sequence
sets could be built. However, if we consider the rate at which
mutations can be fixed by selection, the estimated mutation rates,
and the time since the proposed divergence of human and ape lines,
I think a case can be made to show that humans and apes exhibit
far greater sequence divergence than could be attributed to selection
alone.
>(By the way, the same argument applied here to similar sequences under
>common selection is equally applicable to similar cladistic trees of
>molecules which are interdependent in each organism. In this case,
>independence would have to be demonstrated.)
When I think about it, the same argument could be used to suggest that
there is no direct evidence that the people who claim to be my mother
and father really are my parents (aliens could have implanted an
embryo in my mother instead) or that humans share a common origin.
Perhaps the native people of Australia were "planted" there
separately. Just because Australians can mate with other humans
does not mean that they could not be products of convergence or separate
creations. For the former case, convergence, there is a finite if
astronomically unlikely possibility of occurring. I do not consider the
"ultra-selectionist" arguments for convergent similarities to be terribly
convincing in the case of human/ape sequence comparisons (for other
individual cases perhaps, but not at the level of entire genomes or
chromosomes). For the latter, alien seeding experiments and the like,
there is simply no way of assessing relative probabilities. Given that
there is no auxiliary evidence of an "intrusive intelligence" - eg.
landing field burn marks, or artifacts like titanium tablets dating
from that time, & etc. - at work on Earth during the period I will
opt for an economy of explanations and ignore this possibility
for now.
On the subjects of molecular evolution and sequence analyses, I've
found these newsgroups to be exceptionally informative (perhaps not
as good as chatting in the office of someone who does this work, but
for the internet, these are pretty good sources):
bionet.molbio.evolution
bionet.molbio.info-theory
(see also: sci.bio.evolution and sci.bio.systematics)
http://groups.google.com/groups?hl=en&group=bionet.molbio.evolution
From the bionet.molbio.evolution charter:
"Bionet.molbio.evolution is a forum for scientific discussions and a
source of information for the community of scientists interested in
the study of the processes of how DNA, RNA, proteins and organisms
have evolved at a molecular level. [...]"
Those groups are visited be people who could better address your
question. In fact, I suspect a search of past threads at www.deja.com
might find the question already answered.
Regards,
Tim Ikeda (tikeda@sprintmail.com)
This archive was generated by hypermail 2b29 : Sun Mar 24 2002 - 12:59:04 EST