Re: Crossing a Vast "Neutral Gap" in One Step


hersheyhv wrote:

> > > Apparently by "template matching" you mean those modifications of
> > > pre-existing structures that change binding affinity for a substrate or
> > > co-factor so as to increase affinity.
> >
> > Not at all. By template matching, I mean functions that show an
> > increase in selectability with increased binding of one protein to
> > another as the only change needed. No enzymatic activity is realized,
> > no alterations are needed to either structure, nothing at all is needed
> > other than increased binding to a pre-established protein structure.
>
> IOW, this template matching is not due to mutational change?

Where did you get that from what I just said? Of course template
matching is due to mutational changes. It is just that the no other
function except increased binding is needed for positive selection.

< snip >

> > That is template matching and that sort of evolution can happen very
> > quickly. Examples of template matching functions include the improved
> > evolution of phage-pilus binding and improved antibody-antigen binding.
>
> Both of these involve mutation and selection, Sean, for modified
> binding affinity.

Of course they do. I never said otherwise.

< snip >

> > Exactly - still very easy to achieve. For all types of bacteria such
> > resistances could be gained in just one or two generations given a
> > decent population size. This is not true for the rate of evolution of
> > a novel enzymatic-type function. Only rarely will certain specific
> > types of bacteria have what it takes to evolve a novel enzyme at all,
> > even given tens of thousands of generations of time. See the
> > difference?
>
> Yes. I recommend that you read the following:
>
> http://aac.asm.org/cgi/reprint/39/6/1211?ijkey=583311b537d5e884f0c99a92af0039d11cd064d7&keytype2=tf_ipsecsha
> >
> The key point is right up front:
>
> "b-Lactamases (EC 3.5.2.6) have been designated by the Nomenclature
> Committee of the International Union of Biochemistry
> as ''enzymes hydrolysing amides, amidines and other
> CON bonds . . . separated on the basis of the substrate: . . .
> cyclic amides'' (323). [elipses are in original] These enzymes are
> the major cause of
> bacterial resistance to b-lactam antibiotics and have been the
> subject of extensive microbiological, biochemical, and genetic
> investigations. Investigators have described more than 190
> unique bacterial proteins with the ability to interact with the
> variety of b-lactam-containing molecules that can serve as substrates
> or inhibitors (45, 46, 129, 184; this minireview). Because
> of the diversity of enzymatic characteristics of the b-lactamases,
> many attempts have been made to categorize these enzymes
> by using their biochemical attributes."
>
> That hardly sounds like it is impossible to generate functional
> beta-lactamases.

It's not impossible. It's just that the evolution of the penicillinase
function has never been observed. Every time a bacterial colony gains
penicillin resistance via penicillinase production, the penicillinase
gene was either already there before exposure to penicillin or obtained
via lateral transfer. This penicillinase enzyme has simply never been
observed to evolve.

> > > Typically resistance that involves a new
> > > function really means that an enzyme that originally could only, say,
> > > cleave the toxin at a level below that which effectively removed it
> > > from the cell now is *modified* so that it cleaves the toxin at a
> > > higher rate (often at the expense of its original function to some
> > > extent or other).
> >
> > That's not a "new" function. That's an improvement of a function that
> > already existed in the gene pool.
>
> Hmmmm. So does that mean that any *minor* insignificant (at a level
> that would have selective value) binding or protein-protein interaction
> tthat *could* conceivably be improved on by mutation is not a "new"
> function? I am quite serious in asking what constitutes a "new"
> function. Most proteins interact at some level with other proteins.
> Most biologically relevant molecules have enough similarity that they
> interact with some protein at some (usually selectively insignificant)
> level.

If the level of a particular type of protein function, like a certain
type of motility or enzymatic activity, is below that which would
provide a selectable advantage, then it's gain of a selectable level of
activity would make it a "new" function.

> > > Another way to accomplish this, of course, is simply
> > > to *increase* the amount of enzyme made and either simply sop up all
> > > the antibiotic (as in methotrexate resistance) or produce enough so
> > > that even the low efficiency of cleavage is adequate to reduce toxicity
> > > significantly.
> >
> > Again, that's not a "new" function.
>
> At a selectively relevant level, it *is* a new function.

Yes, but until that selectable level is reached, natural selection
cannot guide the stepwise increase in production or activity.

> > > And lactase, as has been repeatedly pointed out, is
> > > only harder to achieve if you intentionally destroy ebg beforehand. If
> > > you don't lactase function is no more difficult to achieve than any of
> > > your hypothetical template matching phenomena.
> >
> > Lactase is more difficult to achieve for the *average* genetic sequence
>
> Cells do not evolve enzymes from any sort of "average" genetic
> sequence. They only evolve enzymes with new functions by modification
> of one or more of the 10^13 functional genes that already exist in
> organisms.

Yes, but you have to figure the odds that a sequence that is already
very close to producing a selectable lactase already exists in the
genome of a given bacterium that does not already have this functional
ability. The odds for a gaining resistance via antibiotic-target
disruption are very high. The odds for gaining resistance via
production of an anti-antibiotic enzyme, like penicillinase are
relatively low.

> I guess you would not call ebg to lactase the evolution of
> a new function in any case, because the ebg already has affinity for
> lactose and probably already has a small insignificant capacity to
> increase the rate of cleavage of the b-galactosidase linkage.

As I've told you many times before, I do accept ebg to lactase as the
real evolution of a novel function since the level of ebg lactase
before the mutation was not at a selectable level.

> Which brings you back to claiming that the only evolution you will
> accept as being "real" is that which starts with a sequence *maximally*
> distant from a teleologic current function and the unevidenced
> assertion that the only way to evolve that function is by a long random
> selectively neutral walk. I.e., strawman.

This is your strawman version of my position. I've corrected you on
this many times now, yet you keep asserting this strawman like I've
never said anything contrary to it. Again, I never said that the
starting point must be maximally distance for "real" evolution to
occur. As I've said over and over again, you can start wherever you
want. I'm talking about average times to success at a given level of
complexity. Even if you happen to start one point mutation away from a
particular high-level function, the average time to gain other
functions at the same level, will not change. The average time will
still be trillions upon trillions of years for functions, say, at the
level of a flagellar motility system.

> > - which is not true for the average genetic sequence to achieve forms
> > of antibiotic resistance where the goal is the disruption of a
> > pre-established antibiotic-target interaction. Lactase resistance in
> > E-coli is dependent on certain genetic sequences being there, like ebg.
>
> Well, duh. You have to argue that it is impossible for there to be a
> structure like ebg in any real organism.

I never have. I've argued that is it more and more unlikely as you
move up the ladder of complexity. The lactase function, though much
higher that many forms of antibiotic resistance, is still quite low on
the ladder - involving a minimum of no more than 400 fairly specified
residues.

> > Remove these, and lactase resistance will not evolve back at all -
> > even given tens of thousands of generations of time.
>
> Even this is not true, since lactase can evolve from other structures,
> including antibody structures, with the latter occurring in a human
> lifetime.

What are the odds that a particular bacterial colony, like E. coli,
will have access to these other structure? Not very good. That's the
problem. Real time experiments, such as those done by Barry Hall, show
that the evolution of the lactase function is quite a challenge for
most bacteria - even with genomes over 4Mbp and large colonies over the
course of tens of thousands of generations of time. This lactase
function is clearly at a much higher level of complexity than
antibiotic resistance that arises via alteration of antibiotic-target
interactions.

> > This is not true
> > of those forms of antibiotic resistance that are based an disruption of
> > the antibiotic-target interaction. Regardless of what you
> > change/delete/add in the genomes of these bacteria, such resistance
> > will evolve very quickly - usually in less than a handful of
> > generations.
>
> So it all depends on the existence or not of particular *structures*
> and whether or not those *structures* can be modified to produce the
> appropriate *function* in a few mutational steps. That is exactly the
> point I have been making.

Don't you understand that a loss of a pre-existing function/interaction
is much easier to achieve than the gain of a novel function that is not
based on messing something else up?

< snip >

> > > You don't know that. Whatever ebg does is probably disrupted or
> > > destroyed when that enzyme acquires lactase activity.
> >
> > Not true. There is no evidence that any significant loss to any other
> > function was realized when the ebg sequence gained a single point
> > mutation to realize the lactase function. Beyond this, compensatory
> > mutations can allow a single protein to do multiple functions at close
> > to wild-type levels.
>
> So? Not all functions are vital, at least under all environmental
> conditions.

The point is that the gain of the lactase function is not based on
destroying some other function. Even if another function were destroyed
in the process, this is not the basis of the creation of the lactase
function. On the other hand, the destruction of a pre-established
functional system/interaction is the basis of certain forms of
antibiotic resistance. Don't you see the difference here?

> > > Because you are pretending that all resistance is due to "disrupted
> > > interaction".
> >
> > No, I'm not. I'm saying that a certain form of antibiotic resistance
> > is indeed based on the disruption of a pre-established
> > antibiotic-target interaction. This is a fact. The use of the term
> > "disrupted" is also used to describe this in scientific literature.
> > You're reaching for straws on this one Howard. Come on now. You can
> > do better than this.
>
> You are *pretending* that somehow a mutation that modifies a protein to
> reduce affinity is less difficult to obtain than one that increase
> affinity.

No, I'm not. A both decreases and increases in affinity, in a stepwise
manner, are pretty much equally easy to achieve as long as such changes
affect a selectable level of function. Changes in the level of lactase
activity, for example, would be easy to achieve once the level of
lactase activity reached a selectable level where nature could actually
see it. If the level of lactase function is below the level that
nature can detect, then a gain or a decrease in lactase activity, below
this threshold would not be under the guidance system of natural
selection. Therefore it would be much harder to get above this
threshold if lactase would be beneficial than to get below this
threshold if lactase happened to become detrimental.

> > When resistance is based on the creation of an enzymatic function, like
> > penicillinase, then you are wrong. The evolution of such an enzyme
> > would indeed require the crossing of sizable gaps that would indeed
> > slow evolution down a great deal in comparison to "Breaking Humpty
> > Dumpty" forms of antibiotic resistance. In fact, penicillinase has
> > never been shown to evolve in real time. Any time a bacterial colony
> > gains penicillin resistance with the use of penicillinase production,
> > it is because that colony always had the penicillinase gene or because
> > it obtained this gene, preformed, through horizontal transfer. The
> > penicillinase function simply never evolves in real time.
>
> Penicillinase often evolves from penicillin-binding proteins, which are
> the very proteins that causes penicillin to be toxic.

Not true. Penicillinase has never evolved in real time. If you can
prove me wrong, I'd be very interested in the paper documenting the
actual evolution of the penicillinase function in a colony of bacteria
that did not already have the penicillinase gene.

> > I never said that all resistance is based on disrupted interactions
> > Howard - and you know it.
>
> No. But you tried to imply it. And only talked about other
> possibilities when you were called on it.

You've gotta be kidding!? I tried to imply it?! You know this isn't
true. Why keep building these strawman misrepresentations so
deliberately?

> > > Not as a vague
> > > nonsense hope that all resistance is due to "loss of information".
> >
> > I never said that Howard. I said that some forms of antibiotic
> > resistance are indeed due to a loss of a pre-established interaction -
> > and they ARE! Geez! Have you lost your mind?! Wait . . . . . I know
> > the answer to this one already ; )

____________

> > > No. You have *inaccurately* claimed that function (which is an
> > > activity) can be quantitated by a measurement of size. That is wrong.
> > > It is wrong with cytochrome c. It is wrong with lactase. It is wrong
> > > with strep resistance. It is wrong with eubacterial flagella. It is
> > > just plain wrong. That you keep repeating it without trying to justify
> > > your assertions doesn't make them true.
> >
> > Show me how to build any one of these systems with significantly fewer
> > residues than I've listed for them.
>
> *Function* requires structure, but it is not the same as structure.
> You cannot measure *function* by size. You measure function by
> measures of *activity*, not size. To do otherwise is to lie about the
> meaning of basic words like "function".

Again, how do you build a structure with a selectably beneficial level
of flagellar motility with significantly less than 30,000bp? How do
you build a lactase "structure" with significantly less than 1200bp? -
or a cytochrome c with significantly less than 240bp?

You see, all functional systems, with a given type of function, do
indeed require minimum structural sizes and specificities that are
different for different types of functions. And, flagellar motility is
a specific type of motility that is much more complex than some other
types of motility.

> >
> > < snip >
> >
> > > > A
> > > > useful flagellar motility system cannot be built with significantly
> > > > less than 30,000bp of genetic real estate. You argue that this is
> > > > beside the point,
> > >
> > > Rather, I point out the *fact* that size is irrelevant.
> >
> > It is not irrelevant if you can't build such a system with fewer
> > residues. However you say such a system evolves, it still requires this
> > minimum size and specificity. You can say that such levels can be
> > evolved, but you cannot say that such systems are not at much higher
> > levels of complexity than other systems of function.
>
> You can say that the *function* of rotary motility requires
> *structures* that have 1) a rotateable pore with a whip, and 2) a
> motor.

That's exactly what I'm saying and what you seem to be arguing against.
You have to have this minimum in place before you can gain the function
of flagellar motility. This minimum requirement involves many thousands
of bp of genetic real estate to code for - at minimum.

> But those structures exist or they don't. Their size is
> irrelevant to whether or not they can combine to form the *function* of
> rotary motility.

Now that's a whole different argument. You must admit, fist of all,
that a sizable minimum needs to be in place for the final product, the
flagellar motility system, to actually work. How you get there does
not change this fact.

You say this fact is irrelevant to how you get there, and I say it is
very relevant. Let's first agree, though, that this fact is in fact a
reality.

> The feature that determines whether or not they can
> combine to generate the *function* of rotary motility is the number of
> mutational steps required to link the two subsystems that lack that
> *function*, not their size.

Their size and specificity says a lot about the odds that what is
needed will actually exist in the genome. That's the relevance of the
minimum size and specificity requirements.

>Ascribing the difficulty of generating a
> function on the basis of the size of the subsystems rather than the
> number of mutations needed to generate the function is just plain
> silly.

But the size and specificity requirements say a lot about the odds that
a particular system, like a flagellar motility system, will actually be
within reach of what already exists in a genome that has never been
able to produce flagellar motility before.

> > > > There is nothing mysterious or vague or emotional about this
> > > > definition. I'm simply amazed at the difficulties you dream up in order
> > > > to fain misunderstanding.
> > >
> > > I am not fantasizing that the definition is utterly without utility and
> > > tells us precisely nothing about how any *function* arose.
> >
> > That's a whole different discussion. We aren't talking about that at
> > this point. The point is that the system of flagellar motility is
> > indeed at a very high minimum level of functional complexity.
>
> That is not a fact. That is a nonsense phrase. You have no measure of
> "level of functional complexity" that you can apply with any
> consistency at all. You don't even try. You keep refering to total
> sequence size as if that were a measure of *function*. It is not.

You just admitted that the flagellar motility system requires a certain
minimum portion of genetic real estate with a certain degree of
specificity. I refer to this minimum requirement a level of functional
complexity. Call it whatever you want, this minimum does indeed exist
and it is very different for different types of functions. Functions,
or "functional systems", can indeed be differentiated by such minimum
requirements. They are not all the same.

> > It does
> > require a large number of fairly specified residues.
>
> The *function* requires structures that had other *functions*. The
> number of residues is not a measure of *function*.

It doesn't matter if the subparts of a system have other functions or
not. The system still requires the minimum to be in place before it
begins producing a particular function, like flagellar motility, at all
- even a tiny little bit. The minimum must be there. That's a fact.

> > This is true
> > regardless of how you say this system evolved or didn't evolve. It's an
> > independent fact that exists regardless of its true orgin - via ID or
> > via Darwinian evolution.
>
> What you have to do is demonstrate that the number is somehow
> *relevant* to the mechanism by which the *function* arose.

Yes, I do have to do this. But, first, you have to recognize this fact
for what it is. You cannot deny that this minimum must be in place and
that certain functions, like flagellar motility, require a much higher
minimum than do other functions, like lactase activity.

< snip >

Sean Pitman
www.DetectingDesign.com

.

Relevant Pages

  • Re: Crossing a Vast "Neutral Gap" in One Step
    ... >>> increase in selectability with increased binding of one protein to ... Actually what you seem to be describing is incremental mutation and ...
    (talk.origins)
  • Re: Random genetic changes
    ... If so then i can not see Evolution to be in contrast with Intelligent ... Mutation and other forms of genetic alteration provide the ... The idea that a new protein must be a result of hundreds of little ... Read some of the Talk Origins web site resources or a textbook ...
    (talk.origins)
  • Re: Sean Pitman: definitions wanted
    ... >>> protein is at least reasonable. ... >>> one level or another of sequence space. ... And since in evolution, there ... > that beneficial islands spread out in sequence space (see linked ...
    (talk.origins)
  • Re: The Starting Point Problem - for Howard Hershey
    ... the evolution without DNA, RNA, or ribozymes, happens at all. ... here is how modern cytochrome c might have come into existence. ... "Within one or two residue changes". ... rather than with the mutation. ...
    (talk.origins)
  • Re: Speculative Design Hypothesis (with predictions)
    ... It is the definition of science as it ... So you are asserting that the "intelligent designer" has been ... That's as preposterous as me asking if evolution has been observed ... Let's take the process of protein synthesis for example. ...
    (talk.origins)