RE: [OriginsTalk] Re: What's driving evolution?
- -----Original Message-----
From: gromit54609 [mailto:gromit54609@...]
Sent: Thursday, December 27, 2001 4:47 PM
>Perhaps because such mutations have been observed?Of course mutations occur, but that is not the same as saying they are
the primary source of offspring variability. From everything we know of
evolution, it would seem a given the affect upon the genome that is
generating offspring variety is likewise driving the evolution of
Nature selects from the variations found among offspring, and there is
overwhelming evidence to support that genetic recombination is
responsible for creating these differences. Instead of recognizing that
evolution occurs as a result of changes made during meiosis, they are
insistent that mutations are responsible.
>But I wonder how you can reach the design inference from your proposal?If the genetic manipulations driving evolution are created by reactions
performed by the molecular machinery, then these changes occur by design
if indeed the cell is a construct.
>In this case could you provide us withthe required steps that would allow one to eliminate chance and
Meiotic recombination is not random. You can not randomly exchange
chunks of chromosomes and generate a successful outcome with every
manipulation. However, offspring viability in almost all organisms is
very near 100%. If these events occurred by pure chance, death
frequencies similar to those found when researchers are engaged in
mutagenesis studies should be seen.
Meiotic recombination is responsible for the accumulation of new
alleles, and the evolution of new organisms. These reactions are
performed by the cell to introduce offspring variability through genetic
manipulation, and their ability to alter the genome remains
That's not ala Dembski. Pure Ashcraft....
- Daniel (previously):
> > You haven't done much research on this, have youChris:
> > Chris?
> I am not a topic of this list. There are plenty ofDaniel:
> Creationist-bashing lists available where you can be
> free to attack the knowledge of the list members, but
> here you will not. Post information, but do not refer
> to the list members is this manner.
My apologies. I have an unseemly tendency towards sarcasm,
and occasionally it gets out of hand. I'll do my best to
ensure that it doesn't happen again.
> > If you're interested, there are hundreds of furtherDaniel:
> > examples of sequencing of recombination sites - just
> > say the word and I'll provide them.
Before I go any further, I'd like to make a rather trivial
point - if you edit my quoted words, please indicate that
you have done so by inserting an ellipsis. In this case
the meaning of my statement was not really affected (the
words "...both meiotic recombination proteins and..." were
deleted), but it's a good practice to get into. It avoids
accusations of things being taken out of context and so
> If there are indeed, then you should have no troubleDaniel:
> posting at least a few. The references you provided
> are to verifications of sequenced hot-spots or regions
> where recombinations are known to be frequent. I said,
> not one crossover has been sequenced. The references
> you provided do not support a contention of this
Since many recombination "hot-spots" are *identified* by
sequencing meiotic crossovers, and indeed each "hot-spot"
must in fact *be* the end result of more than one meiotic
crossover event (by definition), the articles I cited
provide exactly the information you're looking for.
Nonetheless, I do have an unambiguous example of sequence
data on recombination events, for your gratification:
The article describes the PCR amplification and sequencing
of a region known to have been involved in meiotic
crossover events in several patients with either of two
genetic diseases. Happy now?
> There is a big deal of difference between knowing theDaniel:
> sequence of a hotspot, and of the subsequent
> crossovers that occurred in that area. Using present
> technologies thousands of crossovers may occur in any
> given region, and still only detectable as a single
> event. We are still almost totally ignorant of what
> reactions are performed during meiotic recombination.
> Gametes must be sequences and compared to the parental
> genome before an accurate picture will begin to
You want *entire genomes* to be sequenced before we can
claim to understand the process? I think we already know
quite a lot about meiotic recombination (certainly a lot
more than you continually imply, as I think I have
thoroughly demonstrated over my last two posts). I also
think we know enough to comment on your hypotheses
regarding the importance of recombination in evolution,
as I will do below.
> Some of the major binding proteins have beenDaniel:
> recognized, and regions of high activity are known but
> our understanding of crossovers does not support the
> claim that these reactions are limited to regions
> between preexisting reading frames. The point was;
> there is a leap to the assumption that these reactions
> are not responsible for the creation of new alleles,
> and therefore the evolution of organisms.
Who is making this assumption? I freely concede that
meiotic recombination can, in some cases, generate new
alleles. I have no doubt that mutations occurring during
this process have played a significant role during
evolution. However, I dispute your claim that meiotic
recombination is the "primary" source of genetic
diversity, and I strongly dispute your implication that
mutations caused during meiosis are non-random with
respect to fitness (if this is indeed what you are
implying - let me know if I'm mistaken).
I've never denied that crossovers can occur within reading
frames - such events are extraordinarily rare, but they
certainly occur. However, as recent recombination mapping
of the yeast genome has unequivocally demonstrated, the
vast majority of recombination events occur *between*
genes, not within them. The sequence motifs associated
with these "hot-spots" (such as di- and trinucleotide
repeats and certain transcription factor binding sites)
have been identified, and the process of recombination
has been at least broadly characterised (see below for
detail). There is nothing in this research which has
suggested meiotic recombination to be either a major
source of genetic variation or non-random.
> We know that meiotic recombination is responsible forDaniel:
> the variations found among offspring from the same
> parents, and likewise the differences found between
> breeds of the same kind of animal.
Certainly not always. Each of us possesses approximately
five brand new mutations which were not present in our
parents, most of which have no real effect on fitness. As
for animal breeds, there are certainly some breeds which
can be chalked down to non-meiotic mutation events. There
are about a zillion mouse and fruitfly breeds with
characterised mutations, for instance.
> However, when evolution is taught, there is no mentionDaniel:
> of the genetic rearrangements performed during meiosis.
> We are taught that random mutations such as copying
> errors, and those introduced by external mutagens are
> responsible for the offspring variations selected by
But Chris, the changes introduced by meiotic recombination
(those other than trivial "reshuffling" events, that is)
*are* "copying errors"! They occur when recombination goes
awry, and the various DNA repair systems fail to spot or
deal with the mistake. In this respect they are precisely
equivalent (indeed indistinguishable) from any other
random "copying error", such as base mismatch, UV-induced
thymine dimerisation repair errors, polymerase slippage
and "loop" amplification events, and so on.
And like any other random mutation, these mutations occur
at a given rate irrespective of their phenotypic
consequences. They are, and this bears repeating for
emphasis, entirely random with respect to fitness. There
is no evidence that meiotic recombination errors, or any
other process of mutational change for that matter, occur
at different rates depending on their fitness effects.
Meiotic recombination errors are not usually specifically
discussed in high-school evolution classes for the same
reason that the mechanisms of base misincorporation or
polymerase slippage are not discussed - because they are
complex mechanisms which children don't need to understand
in order to understand evolution. Since the outcomes of
all these processes are the same - random changes to DNA
sequence - the mechanisms themselves are irrelevant.
> This illustrates a contradiction between the knowledgeDaniel:
> we possess and that which is taught. It is more likely
> genetic recombination is the primary source of new
> population alleles, and evolution occurs largely as a
> history of recombination and selection. Mutations
> certainly have an effect on the evolutionary history,
> but they are not the principal agent introducing the
> variability being selected. Offspring varieties is
> first and foremost produced by genetic recombination,
> and it is from these differences that nature selects.
OK, so you keep asserting. But where is the evidence for
this, Chris? Where are the data that support this claim?
Seriously, if this is so "obvious" (as you claim below)
there should be mountains of evidence supporting your
claims - so where is it?
> Proteins involved in the formation of the synaptonemalDaniel:
> complex have indeed been isolated, but have any
> proteins been characterized yet that are actually
> involved in performing a meiotic crossover?
Certainly. I'm most familiar with meiotic recombination in
yeast, so I'll use the process in yeast to demonstrate
what we know about these proteins.
There's a great article about this in the journal
Genetics, but I think you might need a subscription to get
access (ah, the joys of being attached to a university!).
In case you want to hunt the article down, here's the
Kearney, H.M., Kirkpatrick, D.T., Gerton, J.L., and Petes,
T.D. (2001). Meiotic Recombination Involving Heterozygous
Large Insertions in Saccharomyces cerevisiae: Formation
and Repair of Large, Unpaired DNA Loops. Genetics 158:
Now, I'll start at the beginning. The most well-
characterised protein in meiotic crossover events is
Spo11p, which is a homologue of the DNA super-coiling
protein topoisomerase II (TIII) and, like TIII, catalyses
the formation of double-strand breaks (DSBs). I don't know
how much you know about the biochemistry of recombination,
but these DSBs are required for the process to begin. After
the DSB has been formed the 5' ends of the break are
resected (that is, the two strands are pulled apart) by a
cluster of proteins called the Rad50p/Mre11p/Xrs2p complex
to form two long single-stranded DNA (ssDNA) "tails".
It starts getting a little theoretical here, as some of
the processes involved in rejoining the strands are
controversial (and somewhat difficult to study, although
not by any means impossible). The most popular current
model (and the most likely one, IMO) involves the
"invasion" of the homologous chromosome by one of these
"tails", creation of a heteroduplex DNA molecule (a little
hard to explain without diagrams), followed by ligation
and cleavage to produce two separate, recombined
(As an aside - the heteroduplex structures have been
directly observed in meiotic yeast cells; see: Allers, T.,
and Lichten, M. 2001. Intermediates of yeast meiotic
recombination contain heteroduplex DNA. Molecular Cell.
The proteins involved in the actual rejoining of the
strands definitely include Dmc1 (a Rad51 family member, in
case you're familiar with DNA repair proteins):
"Dmc1 is highly similar to Rad51 in both sequence and
function (both possess strand-exchange activity), but is
specifically required in meiotic recombination."
Certain mismatch repair proteins are also involved here,
including Msh4p, Msh5p, Mlh1p, and Mlh3p. And there's more,
and more, but I'm supposed to be doing my own research
right now so I'll leave the rest of it up to you. :-)
Anyway, I've certainly given you some clear examples of
proteins involved in meiotic crossover events. I encourage
you to check out the references I've supplied, find some
more of your own, and get back to me if you have any
> Is there any theoretic or experimental justificationDaniel:
> to propose that a crossover could not make any manner
> of genetic exchange including replacing single
> nucleotides, codons, control regions, etc.?
It would have a bugger of a time trying to introduce a
small inversion mutation (in fact I have a sneaking
suspicion it couldn't do it at all). It *could* introduce
point mutations, but with some difficulty (especially when
you consider the extensive mechanisms present to repair
any mismatch mutations occurring during meiosis - there
are at least four mismatch repair (MMR) proteins known to
be involved in the process of meiotic recombination).
Duplications and deletions, of course, it could do with
relative impunity, and chromosomal translocations are
possible with any process that involves a DSB.
But the *possibility* of such things occurring really isn't
the issue, is it? Your assertion is that meiotic
recombination is the *primary* source of such events, so
the important question is how often these things occur. So,
is there any empirical reason to expect that such events
are the "primary" source of new alleles, given that *all* of
these processes occur with much higher frequency outside
the meiotic recombination pathway?
Look, I have no problem with the *idea* that most genetic
changes occur during meiotic recombination. It doesn't
clash with any preconceptions I might hold. It doesn't
trouble me as a scientist. If an article was published in
Nature tomorrow demonstrating your claim to be true, I'd
accept it without undue scepticism. But I haven't seen a
single scrap of hard evidence to support this idea, and
until I do I'll continue to regard it as nothing more
than unsupported speculation.
> It seems obvious that evolution occurs as a result ofDaniel:
> the alterations made during meiotic recombination.
Chris, seriously, by what definition of the term is this
hypothesis "obvious"? I can respect it as a tentative
hypothesis awaiting empirical support, and, if you are
interested, I can even help you look for support for it.
But at this stage it is a speculative theory, nothing
more, nothing less.
> Why is there an adamant adherence to mutations?Daniel:
What do you mean? Changes occurring during meiosis are
still mutations, Chris. *Any* change to a DNA sequence is,
by definition, a mutation.
> It seems a refusal to recognize the design inferenceDaniel:
> behind evolution as is demonstrated by the fact that
> internal mechanisms are producing the variations from
> which nature selects.
Actually, all evolutionists believe that "internal
mechanisms are producing the variations from which nature
selects". They're called mutations, and they are random
with respect to fitness. This in no way represents any
form of evidence for "the design inference", so I'll
assume you were actually trying to intimate that meiotic
recombination represents a mechanism for generating non-
Now, I'll try to say this as politely as possible, Chris,
and please try not to take it the wrong way. This
argument of yours is astoundingly tenuous, and frankly
will need a *lot* of work if it is to be anything more
than idle speculation.
Firstly, you need to demonstrate that meiotic
recombination events are the "primary" source of new
alleles. This means taking a close look at the processes
of recombination. How likely are these processes to cause
mutations of specific types? How frequently would we
expect these mutations to occur? How well do these
frequencies compare to the observed frequencies caused by
other mutational mechanisms?
Secondly, you need to demonstrate that meiotic
recombination produces non-random changes with respect to
fitness. It is not enough that recombination tends to be
site-specific (since this also clearly applies to most
other forms of mutation). You have to show that meiotic
mutations are more likely to be phenotypically beneficial
than other sorts of mutation. Then, and only then, can you
convincingly claim that meiotic crossover represents a
non-random mechanism for generating genetic change.
Thirdly, and this only applies if the first two claims
have been supported, you need to show how such a process
is necessarily evidence for "the design inference". Non-
random mutations do not necessarily imply design - in
fact, the recent controversy over "directed mutation" was
sparked by an ardent evolutionist. It is possible (but
extremely unlikely, for several reasons) that evolution
has "devised" a way to direct mutations to achieve
specific phenotypic goals. I agree that this is unlikely,
but it is certainly a possibility you will need to rule
out before you can claim decisive support for your
I hope all this is helpful, anyway. Your idea is an
interesting one, and quite possibly worth pursuing (if
nothing else you would learn quite a lot about genetics
and biochemistry while researching it), but it needs a
whole lot more work. If you get stuck for references or
articles let me know and I'll see what I can find. Oh,
and any comments on the above post would be welcome
considering how long it's taken me to write it. :-)
http://my.yahoo.com.au - My Yahoo!
- It's My Yahoo! Get your own!
> Of course mutations occur, but that is not the same asDaniel:
> saying they are the primary source of offspring
> variability. From everything we know of evolution, it
> would seem a given the affect upon the genome that is
> generating offspring variety is likewise driving the
> evolution of organisms.
> Nature selects from the variations found among
> offspring, and there is overwhelming evidence to support
> that genetic recombination is responsible for creating
> these differences. Instead of recognizing that evolution
> occurs as a result of changes made during meiosis, they
> are insistent that mutations are responsible.
*sound of lightbulb going "pink!"* I think I'm finally
beginning to understand where you're coming from here. Let
me try to outline your basic argument, and tell me if I've
got it right.
1. The primary source of offspring variability is due to
2. Evolution functions by selecting amongst offspring
variability for the most highly adapted individual.
3. Therefore meiotic recombination must be the primary
source of variability for evolution.
Is that it? Feel free to change it if I haven't quite
captured the essence. Once I've figured out exactly
what it is you're arguing, I'll address it. :-)
> > But I wonder how you can reach the design inferenceChris:
> > from your proposal?
> If the genetic manipulations driving evolution areDaniel:
> created by reactions performed by the molecular
> machinery, then these changes occur by design if indeed
> the cell is a construct.
Erm...this is actually a rather blatantly circular
argument, Chris. You're saying that "these changes occur
by design if indeed the cell is a construct", and then
using this premise (which is predicated on the assumption
of design) to support your "design inference"! Perhaps
you should rethink this argument? :-)
> > In this case could you provide us with the requiredChris:
> > steps that would allow one to eliminate chance and
> > regularity?
> Meiotic recombination is not random. You can notDaniel:
> randomly exchange chunks of chromosomes and generate a
> successful outcome with every manipulation. However,
> offspring viability in almost all organisms is very
> near 100%. If these events occurred by pure chance,
> death frequencies similar to those found when
> researchers are engaged in mutagenesis studies should
> be seen.
Indeed. I think you have convincingly established (not
that it was ever in doubt) that meiotic recombination is
not a random process. That eliminates chance as an
explanation for this phenomenon. But according to
Dembski's design "filter" (which gromit is, I assume,
referring to), the design inference is made by first
eliminating the possibility of the event in question
occurring by either chance or *regularity*. You have
failed to address the possibility that meiotic
recombination can in fact be explained with reference to
naturally occurring processes.
I suggest that it can easily be explained by reference to
perfectly natural processes. Meiotic recombination is non-
random because of the nature of the mechanism, which
involves template-directed recombination triggered by a
series of sequence-specific protein binding events. The
mechanisms which regulate it are the product of another
natural process - the process of evolution, which would
rapidly eliminate any organism incapable of regulating its
meiotic crossover events.
> Meiotic recombination is responsible for theDaniel:
> accumulation of new alleles, and the evolution of new
> organisms. These reactions are performed by the cell to
> introduce offspring variability through genetic
> manipulation, and their ability to alter the genome
> remains theoretically limitless...
Interesting. You are in fact engaging in yet another
circular argument here, although this one is more subtle.
You are ascribing teleology to the cell (by stating that
recombination is "performed by the cell to introduce
offspring variability through genetic manipulation") and
then using this claim to support your inference of
design. This claim is also a non sequitur; you are
taking a non-controversial premise (that meiotic
recombination plays a role in generating new alleles) and
wildly extrapolating it to support your conclusion (that
the cell must have been designed).
But I've pointed out the problems in your argument in a
previous post, so I'll leave it at that for now. Let me
know what you think about my summation of your argument,
and we can go from there.
http://my.yahoo.com.au - My Yahoo!
- It's My Yahoo! Get your own!