Re: [asa] Anybody familiar with Sanford's book?

"near-neutral (very slightly negative)"

Near-neutral are about as likely to be slightly positive as slightly
negative, though "natural selection acts on the entire genome as a
whole, not individual nucleotides" is in fact a major problem for
labeling near-neutral, or even many more dramatic mutations, as
positive or negative-we can easily tell, e.g., that a mutation makes
an enzyme a bit more or less efficient, but determining the overall
effect on the well-being of the organism (if any) is much more
problematic-too efficient an enzyme could throw other things out of
balance. The other major problem with various antievolutionary claims
that mutations are usually negative is that negative or positive is
almost entirely situation-dependent. Something that always kills you
before you can reproduce is guarenteed not to be beneficial, but it
also eliminates itself form the populaiton automatically. However,
sickle-cell anemia, near lethal in homozygotes, provides significant
resistance to malaria in the heterozygote-a seemingly bad gene
actually turns out to benefit in the particular set of pressures.
Mutations that are strongly positive or negative have correspondingly
much higher probabilities of spread or elimination in a population.
Not only does the environment vary, but there are different possible
scenarios within an organism. If a gene is just beginning to take on
a new function or a genome is encountering novel conditions, the
probability of a mutation generating significant improvements is
relatively high, as is the probability that the organism can get by
with a poorly functioning version. If a gene has long functioned in
its present role, it's probably fairly well refined and subject to
only minor improvement, and it may be relatively imortant to an
organism.

"rarely can negative mutations be separated from positive ones"

Misleading. ALL negative mutations are difficult to separate from ALL
positive mutations, but any given mutation sorts independently of any
other unless it is fairly close to another on the same chromosome.
Thus, in most cases it's fairly easy to separate them. There are
examples of such linkages, however. For example, strong jaws and bad
backs are associated in the dog/bear group of carnivores, specifically
in pandas and bulldogs.

> Since it has been demonstrated that more negative mutations occur than 'positive' ones (within our lifetimes; disease, etc)<

I don't know of any such thing actually being demonstrated, though
it's popularly asserted in attacks on evolution.

> Because near-neutral mutations cannot be selected against...they will build up
> and their additive affects will eventually condemn the genome to extinction.

No, because if the additive effects are significant (of which there is
no guarentee), selection against the net result will become higher.

> This is an inescapable fact just like if you kept introducing typos into a
> book – the book will be understandable for a long time but eventually it
> will lose sufficient meaning to convey the message (extinction).

However, A) natural selection works to correct typos and B) numerous
different books are all viable options. The result won't be perfect
copies of the original, but if the copies were all perfect there
wouldn't be any evolution.

> No mutation can claim to be truly neutral, as it takes up space in the
> genome – spacing itself (e.g. between regulating elements and their genes)
> is important.

Well, in theory any given mutation puts an organism one step closer to
evolving certain functions and one step further from evolving certain
functions. Similarly, there are factors that can favor a more
G/C-rich versus A/T rich genome. If the potential for future
evolution is discounted as a factor, however, there are plenty of
mutations that have absolutely no effect on the organism. For
example, although spacing is of some importance, the exact length of a
space is not, nor the exact sequence of bases. Likewise, hundreds of
bird species have mitochondria that do exactly the same thing but with
slightly different gene sequences; similar patterns are seen in all
types of organisms but that's one of the largest single data sets.

> high mutation rate per individual (up to 100-300 per person per gen), there is no way enough 'negative-mutant' individuals can be removed/out-competed from the population <

In reality the vast majority of these mutations rapidly disappear from
the population (at least as direct decendants; they may reappear as
new mutations in other individuals) due to genetic drift. Especially
at the beginning, a mutation is present in such low numbers that it
has a very high chance of not being passed on. For example, at the
moment any new mutation of mine has a 50% chance of being carried by
Timothy. If he has children, any one of them has a 25% chance of
having my mutation, etc. Someone with significant negative mutations
would have to reproduce a lot to make much difference.

>Sanford argues that whilst there may be short term 'benefits' in some
mutations...these are only ever losses of already existing information
(turning-off of a gene, etc).<

No, they are also production of new information.

> However, firstly, wherever the free copy is inserted it will instantly
> affect something else (reduce information) in the genome by simple
> positioning (even if quite small effect – it will never have zero effect).

Affecting something else is not reducing information, it's increasing
it. However, there's no guarentee that there will be a significant
effect on anything else. In particular, major methods of gene
duplication tend to generate either copies adjacent to each other or
copies of the entire genome (with things positioned as they were).
Duplication associated with transposable elements is the main
exception.

> Secondly, the amount of that gene product would be instantly increased (the
> free copy would also have to be 'on' in order to be selected for an
> alternative function) leading to an imbalance in gene product and therefore
> function; this could have a profound impact on other genes and/or pathways –
> it will very unlikely have no impact.

Depends on A) what signaling codes are associated with the "new" copy
and B) what function the gene has. A recent study on human digestive
enzymes found one that had been copied several times, improving our
ability to digest certain foods.

> Thirdly, which copy is to become the 'back-up'? Any recombination-exchange
> between the two copies over time will likely contaminate each other with
> their own mutations – actually quickening decay.

It depends greatly on the details of function. Most likely, the basic
structure of both gene products will be similar and the same region
will be functional. As a result, exchanges between most parts of the
gene will have little effect on either, whereas exchange of the
functional regions would merely swap which gene was doing what. Only
recombination that disrupted the functional regions would be a
problem. However, recombination disrupting the functional region
would become much less likely as the functional regions diverged,
whereas if the functional regions were not very divergent,
recombination within them would be likely not to have much effect.
Note also some recent studies observing evidence of such processes.
It's especially easy if the original protein had dual function-each
copy can specialize on one of the original functions.

> Fourthly, genetic drift of the free copy could easily lead to a dominant
> negative affect on the 'back-up' (i.e. the new 'free' version will interfere
> with the originals' function).

? Not quite sure what is being claimed here. I think what's
envisioned is the divergent copy becoming a dysfunctional allele of
the old version. If so, it's simply a repeat of the recombination
point made above.

> Fifthly, how would the first several mutations in the free copy be selected
> for – this is irreducible complexity at the most fundamental level –
> nucleotides function within the context of the surrounding nucleotides,
> which obviously are in place for their original function – not anything new
> (makes it more likely to become a dominant negative mutant).

As the original was functional, no mutations are necessary to give
function to the copy. Starting with a functional original, there's a
good chance that a modification of it will give something with
slightly different function. The idea that several coordinated
mutations must take place is incorrect. However, because there is the
original, the "free" copy is free to become non-functional in the
process of developing a new function. This is also addressed by the
possibility that the original was multifunctional (or at least covered
both functions, perhaps by doing one generic action).

> Sanford makes a nice point in the book about probability and mega-mutations.
> I think it's Dawkins who coined the phrase 'Mount Improbable' for the theory
> of evolution. If one is climbing such a mountain, and is prone to 'tripping
> over' (mutation) occasionally it seems logical that you'd rarely trip-up the
> mountain very far. On the other hand, tripping in a downward direction seems
> more likely, and it's possible to fall a great deal further!

Entirely dependent on where one is on the mountain, as well as what
are included under "mega-mutations". For that matter, a landscape is
much better as a picture of evolution than a mountain-from some
places, one can only fall down; from others, up is more likely. Add
the caveat that the landscape is constantly shifting (rapidly in some
places, slowly in others).

> One of the implications of the general deterioration of the genome is that
> life spans could quite easily have been in the 100s as per Genesis.

Reproductive output per lifespan, rather than longevity per se, is
what would actually be selected for. There's no reason to assume that
"better" genes would produce greater longevity (cf. Gulliver on the
ambiguous benefits of immortality). Some genes do promote greater
longevity, and some clams can surpass 400 years in age (the current
animal record holder), but there's no particular reason to think that
such is biologically "better".

-- 
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.