Re: Blog: The Scars of Evolution.

On Mar 20, 12:21 pm, Treus <treusd...@xxxxxxxxx> wrote:
hersheyh wrote:
On Mar 19, 3:14 pm, Treus <treusd...@xxxxxxxxx> wrote:
hersheyh wrote:
On Mar 18, 6:01 pm, Treus <treusd...@xxxxxxxxx> wrote:
hersheyh wrote:

If the fixation
process alone had ever been shown to transform species S1 to species
S2

You didn't ask me to demonstrate every single change that ever
occurred and when it occurred. You asked if there was a known process
or mechanism that was *sufficient* to account for the observational
difference.

I think you and I can agree that humans and chimpanzees are different
species. And also that any differences that exist between these
species is a consequence of sequence differences in their genomes.

If I can present a known mechanism that is *sufficient* to account for
the sequence differences between these two species' genomes, that is
all I have to do to refute your claim that there is no such
mechanism. In fact, I did more than that. I said that the *slowest*
known mechanism, and one that, because it occurs completely by chance,
cannot be prevented or stopped, the completely chance fixation of
mutations that have no selective consequence is *sufficient* to
account for the observed amount of difference.

Okay, so far you have specified a mechanism, but you have not
demonstrated its _sufficiency_ to accomplish what is being claimed for
it (i.e. common descent). On one hand, we an observed (micro-)
mechanism. On the other hand, we have a claimed (macro-) phenomenon.
Where is the data connecting the two such that the first would be
sufficient for the second?

If you accept that chimpanzees and humans are different species, then
any differences between them that are 'genetic' and thus relevant to
evolutionary differences will be reflected in the observed sequence
differences. You do understand that, don't you? Thus, when I say
that I have a mechanism that *suffices* to explain *ALL* of the
genetic sequence differences we observe between humans and chimpanzees
and that that mechanism is the slowest known mechanism, I *am* saying
that this mechanism can account for *ALL* of the differences between
humans and chimps. Isn't that more convincing and complete evidence
than positing a mechanism that can only explain one or two features
that differs between these two species?

Sure, though you have to go beyond merely _claiming_ sufficiency in
either case, of course.

All you asked was whether there is *a* mechanism that would be
"sufficient" to accomplish what evolution claims. My statement says
that the *slowest* known mechanism for changing genomes is sufficient
to explain *all* the differences between humans and chimpanzees. I
also point out that the assumption that *all* the genome changes were
selectively neutral is obviously incorrect and mentioned 50 sequences
that would have required selection.

Perhaps you are claiming that even though the slowest known mechanism
may be sufficient for most of the differences between chimps and
humans, there are certain specific features (macroevolutionary ones,
by whatever you think macroevolutionary means) that cannot be due to
even the much faster mechanism of selection. My reply is that I
simply do not know of any such feature. Can you point out such a
feature (a protein or DNA sequence) in the genomes of these two
species for which there would be *insufficient* time to account for
the observed sequence differences? There are a very few sequences (50
or so) that appear to have more differences than would likely be due
to chance drift and fixation alone. But all of these few can easily
be attributed to the much more rapid process of differential
selection.

I cannot evaluate whether or not there are any differences between
chimps and humans that *could not* arise by known mechanisms of genome
change until you can point such a gene sequence out for me. The
genomes of humans and chimps are both mostly done by now. So if there
are such sequences, you should be able to find them.

In fact, most proteins (over 99%) are identical (or only differ at a
single non-selectively relevant aa) in humans and chimps, so I would
have a hard time choosing any single protein that would satisfy your
desire that they be *different* enough. The 50 or so sequences
(that's the *population* total out of all the genome, not a *sample*)
that have been found to be so changed that they are unlikely to have
changed due to chance alone (i.e., are likely to have *changed*
because of non-conservative selection) are not coding sequences. They
are most likely regulatory sequences, and regulatory sequences can
change quantitatively.

or feature F1 to feature F2,

The feature I am looking at is the entire genome of these two
species. That pretty much includes *all* the features that differ
between them.

I am asking, essentially, "How much time does it take for completely
random processes to produce two genomes this different." If such
completely random processes are *sufficient* to produce genomes this
different in the time available, then selection, which is much more
rapid, could easily produce those differences that are phenotypically
relevant between the species in that same time frame.

Take as much time as you need. That is not a point I wish to argue.

The time available is crucial. The determination of *sufficiency*, at
root, involves the measure of the *rate of change* of whatever
mechanism or process one applies to produce that change multiplied by
the amount of time that mechanism can work. The *rate of change* due
to a mechanism is measurable right here and now. Applying that rate
to the available time tells us if that mechanism can produce the
needed amount of change. That is, it tells us if the mechanism is
*sufficient* to produce the observed amount of change.

Only assuming you can generalize local observations to extrapolate the
effect as much as necessary. Why do you make that assumption?

I do have to make the quite reasonable assumption that *mutation
rates* have not changed dramatically (like entire orders of magnitude)
within the human and chimp lineage. But since mutation rates that are
much faster would be lethal and those that are much slower would
require non-existent repair mechanisms or DNA that is no longer
subject to chemical reactions or errors in replication, I think the
fact that no such large differences exist in any modern mammal would
be support for the contention that the rates of mutation have been at
least reasonably similar (within a factor of 2 or less, and even that
is wildly unlikely) throughout the 10 million years. If you think
that this assumption is unreasonable, you have to present some sort of
evidence to support that.

*If*, for example, you actually had evidence that the universe was
only 6000 years old (which, of course, you don't), then applying the
observed rate of sequence change due to chance alone (let's say it is
about 3% per 10 million years for that short of a period of time
would produce *insufficient* total change and we could declare the
method *insufficient*. OTOH, if the time available from the common
ancestor were 6 million years, the same *rate* of change due to chance
alone would be *sufficient*. If the time available were 60 million
years, we would have to hypothesize that there were mechanisms that
*prevented* change in sequence. Hint as to what such a mechanism that
prevented change would be called: natural selection.

If
you could show that a process could be extended to sufficiently cause
common descent, that would enough without considering how slow or fast
it is.

The argument, my dear boy, *depends* on the mathematically determined
rate of sequence change (fixation) if such change were due to chance
alone. That number, as you would know if you knew something about
population genetics, is related to the rate of mutation, which is
directly measurable.

I would be happy to go dig up the numbers again if I thought you
actually understood the question you asked. However, it is rather
silly for me to dig up the numbers again if you don't even understand
why the *rate of change* and the *amount of time* that rate is
occurring over is important to determining *sufficiency* of the
proposed mechanism to produce the observed *amount* of change. Hint:
The molecular clock for selectively neutral sequences can be a rough
estimate of how much change one should expect.

It's obvious why rate of change would be important. My point is to
grant whatever timeframe your case may need, to move beyond that
variable and concentrate on the unique properties of your imaginary
global mechanism.

The mechanism I am proposing is not "imaginary". It is a mechanism
that only requires that mutation occurs. The math involved in
completely chance fixation of completely chance mutations is as well
worked out as determining the probability of rolling a six is for an
honest die (also a completely chance event).

If you are willing to grant me whatever time frame I need, I could
choose whatever time frame works. I am not willing to grant myself
whatever time frame I need. Science does not work by positing
whatever hypothetical time-frame is needed or generating hypothetical
posited entities (HYPEs) that can do whatever one needs.

I am only willing to grant myself the time frame that evidence
*independent of sequence change rates* permits. In this case, that
would be non-sequence related evidence for the time of the initial
separation of humans and chimpanzees from a common ancestor.
Specifically, that would be fossil evidence as dated by geological
dating methods.

If you have *evidence* for a different time frame for the separation
of humans and chimpanzees from their common ancestor or *evidence*
(again independent of the sequence differences) for some other time
for the first appearance of these species, you have to present it. I
am restricted to the use of the time frame (independently derived)
that science gives me.

.

Relevant Pages

  • Re: For Sean Pitman: Some real nested hierarchies
    ... The sequence is 1002 bases long, ... bonobo sequence and has a different species of gibbon from tas2r38. ... Different forms of the same gene (alleles) ... between the two groups of modern humans and the Neanderthals. ...
    (talk.origins)
  • Re: You know what I really love
    ... >> gene by a linear sequence of the four bases. ... >> Different species have slightly different sequences, ... >> the patterns of similarity, so I'm going to use a different symbolism to ... I say it's because all the genes evolved on the same tree, ...
    (talk.origins)
  • Re: Yet again, human evolution: huh? (DEAN speaks)
    ... >>gene by a linear sequence of the four bases. ... >>Different species have slightly different sequences, ... >>the patterns of similarity, so I'm going to use a different symbolism to ... >>the African apes, ...
    (talk.origins)
  • Re: For Sean Pitman: Some real nested hierarchies
    ... The sequence is 1002 bases long, ... bonobo sequence and has a different species of gibbon from tas2r38. ... Different forms of the same gene (alleles) ... between the two groups of modern humans and the Neanderthals. ...
    (talk.origins)
  • Re: Human Endogenous Retroviruses in the Primate Lineage
    ... I believe you mean "DNA barcoding". ... more matches are sister species in one clade and the new sample is ... this explains why a mitochondrial gene is used ... Plants have too little mitochondrial sequence diversity, ...
    (talk.origins)