The Nature of Mutations II

I have been putting forward such a statement since the first page of this topic. I would argue that by the specific criteria of a mutation being a 'heritable change', in DNA if you will, then the particular source of the change is not relevant. A transgenic organism reintroduced to the wild will be in competition as much as a naturally occurring mutant and can have a change which is just as heritable.The VDJ recombination is not a 'specific' change, if it was specific it would be useless for generating the neccessary variation required. It is arguably a specific class of change, but then so arguably would be mutations/ knockouts cause by transposon insertion. Or would you say that transposon insertion would only be a mutation in a non-experimental situation. Would you say that trisomy 21 is not a mutation because its occurrence is predictable to some extent?The origin of a specific mutation is not a fundamental feature of the nature of mutations. There are a number of ways a particular mutation could come about. The change itself is the fundamental aspect.

I would say that if the change is wrought deliberately
or predictably, then it is not a mutation.The effect of the change has no bearing on the nature of
mutations ... which was the question for this thread.In terms of stimulii, a mutagen doesn't CAUSE a mutation
it make one more likely. In the immune system pathogens
do cause changes.You cannot map mutation 1:1 to heritable change because not
all mutations are heritable.If we are not interested in the specific source, why bother
defining it in the first place?

Allowing that the effect of the change has no bearing on its nature, then neither does the cause. It is the character of the change itself which is important.I would argue that given the right circumstances there is a 1:1 mapping of mutations to heritable changes. Even in a somatic cell the potential for inheritance, within the cell line, is there; otherwise we wouldn't suffer from cancer.The source of the mutation isn't what we have been trying to define, what we have been trying to do is get a specific definition of what a mutation is, not what causes a mutation. And in evolutionary terms the way a particular mutation comes about is considerably less important than the effect of that mutation.As to pathogens causing changes in the immune system, they certainly do, but they don't cause the VDJ recombination, as has already been pointed out.Mutagens frequently do cause the mutation, how else would you characterise ethyl methyl sulfonate adding and alkyl group to a guanine base and causing it to mispair with thymine. The fact that the muatgen doesn't directly change the base from a G to a A doesn't mean it isn't changing the character of the DNA in a way which results in a heritable change. Why would you say this? What difference, qualitative or quantitative is there between and artificial and a spontaneous point mutation?

quote:

---

Allowing that the effect of the change has no bearing on its nature, then neither does the cause. It is the character of the change itself which is important.

​

[..]in evolutionary terms the way a particular mutation comes about is considerably less important than the effect of that mutation.

---

The we don't need to know any more about a mutation than that
it is an unanticipated change between parent(s) and offspring...
but that's heritable variation NOT mutation.If one is to consider any cell-division to produce offspring then
we can include somatic mutations ... but that's not what heritable
variation means in evolutionary terms.Darwin's original concept of heritable variation is targetted
phenotypically. It's about extant traits (even if those are subtle
metabolic differences). Offspring is considered to be whole-organism
descendents, not cell-lines (changes in which die with the
organism and have no evolutionary impact beyond that one
generation).That's why heritable variation isn't 1:1 with mutation.

quote:

---

I would argue that given the right circumstances there is a 1:1 mapping of mutations to heritable changes. Even in a somatic cell the potential for inheritance, within the cell line, is there; otherwise we wouldn't suffer from cancer.

---

Something is either 1:1 or it isn't ... if it's conditional
then it's 1:1c and the conditionality needs to be defined.You want to define mutation to be heritable variation, even
though it's not, but I'm not sure why.

quote:

---

Mutagens frequently do cause the mutation, how else would you characterise ethyl methyl sulfonate adding and alkyl group to a guanine base and causing it to mispair with thymine. The fact that the muatgen doesn't directly change the base from a G to a A doesn't mean it isn't changing the character of the DNA in a way which results in a heritable change.

---

Which guanine base in the genome is affected? All of them,
some of them? Are they always affected?

quote:

---

I would say that if the change is wrought deliberately
or predictably, then it is not a mutation.

​

Why would you say this? What difference, qualitative or quantitative is there between an artificial and a spontaneous point mutation?

---

The difference qualitatively is that 'someone did it for a reason.'If some-one digs a hole in the ground and fills it with water
we do not call it a lake, we call it a man-made lake, a folly,
ro some-such. Part of the characterisation of a mutation is
that it is a natural phenomenon, which means that artificail,
intentionalm, directed genomic changes are not mutations.I'd allow artificial mutation though (like it matters what I
would allow )

quote:

---

The source of the mutation isn't what we have been trying to define, what we have been trying to do is get a specific definition of what a mutation is, not what causes a mutation.

---

The distinction I see is not about causation (exactly), but that
the nature of a mutation includes that it is a natural
rather than directed phenomenon.

You can't say that a somatic mutatiopn can't be a mutation in evolutionary terms, it can. It is simply a very limited population in a very temporary environment, but the evolution seen in a somatic cell line is no less real even if it has a very limited timescale. Obviously a somatic cell line mutation is not heritable to the next generation of the organism it is within but it is heritable within the cell line.If you don't think that a cell line can evolve then presumably you see no merit to work with E. coli or any of the other standard laboratory microorganisms that are used for evolutionary studies. AAAAAAAARRGH!!! I had hoped it was only the creationists who hadn't realised that evolutionary biology has moved on since Darwin, apparently I was wrong.I never said that mutation was heritable phenotypic 'variation', and in fact you didn't use that wording when you were complaining, you said 'heritable change'. A synonymous mutation is still a mutation, it is still a change which is heritable even if it causes no change in heritable phenotype. This has never been my position. This is in fact a blatant misrepresentation of my position.As to my definition being conditional, it isn't. It is you that wants to add all these conditional terms and caveats such as predictable and that it only applies in a germ cell line. The only thing I am saying is that it is a heritable change even if it is not likely to be inherited due to being in a somatic cell which is normally not cycling. By your criterion a mutation in the germ line of an animal wouldn't be a mutation as long as that animal failed to reproduce.As to guanine bases which are affected, obviously that depends on the dose of EMS. You could affect all of them, but what use would that be, the whole organism would die pretty much instantaneously from that sort of dose I should think. My initial point however was that the induced mutations are often of a specific and predictable class, as are the changes seen in VDJ recombination.As to your qualitative difference, it is no difference in the mutation which was my point, there is no way to distinguish between the 2, its provenace is only a qaulitative difference in the causation, not in the mutation.I see absolutely no requirement for the source of a mutation to be natural? Why on earth should there be one?You are saying that if a fly gets a mutation due to being struck by a cosmic ray that counts but the very same mutation caused by a radioactive source in a lab wouldn't? You claim that part of the characterisation of a mutation is that it is a natural phenomenon. I would be interested in any supporting evidence you have. Certainly the many mutant screens done by geneticists would seem to indicate that the vast majority of biological scientists don't draw that distinction.You seem to want a more and more restrictive definition of a mutation; it has to be unpredictable, it has to have a natural origin, it has to be in a germ line cell none of these are inherent in your initial definition of a mutation as a 'copy error' where in fact you only link that to the concept of heritable variation. Perhaps you could furnish us with your new definition of a mutation so we can see just how restrictive it is.

quote:

---

The distinction I see is not about causation (exactly), but that
the nature of a mutation includes that it is a natural
rather than directed phenomenon.

---

What is the difference then, to you, chemically, between a base change induced by exposure to EMS and one that is caused in a skin cell by a mutagen present in the environment? Chernobyl was an accident in a man made construct...the mutations induced in deer in Norway by the fallout are natural, induced, directed, predicted, unpredicted, heritable, not heritable, mutations, not mutations?How is WK's example of a mutation in a cell lineage different from a germline mutation...germ cells are not a type of cell?Can you give me a natural example of an "anticipated" mutation?How can you tell the difference between a deliberate mutation and an accidental one (excluding site directed mutagenesis).

quote:

---

You can't say that a somatic mutatiopn can't be a mutation in evolutionary terms, it can. It is simply a very limited population in a very temporary environment, but the evolution seen in a somatic cell line is no less real even if it has a very limited timescale. Obviously a somatic cell line mutation is not heritable to the next generation of the organism it is within but it is heritable within the cell line.

---

I can say that if I want A somatic mutation is not 'visible' to evolution.Cell-lines do not continue into future generations of a species
population, and so do not contribute to evolution. They are,
therefore, not heritable variations in the evolutionary sense.

quote:

---

If you don't think that a cell line can evolve then presumably you see no merit to work with E. coli or any of the other standard laboratory microorganisms that are used for evolutionary studies.

---

E.Coli (or any other single-celled organism) cannot be considered
in the same way, since there is no germ-line/soma distinction.

quote:

---

Darwin's original concept of heritable variation is targetted
phenotypically.

​

AAAAAAAARRGH!!! I had hoped it was only the creationists who hadn't realised that evolutionary biology has moved on since Darwin, apparently I was wrong.

​

I never said that mutation was heritable phenotypic 'variation', and in fact you didn't use that wording when you were complaining, you said 'heritable change'. A synonymous mutation is still a mutation, it is still a change which is heritable even if it causes no change in heritable phenotype.

---

This was just in reference to your insistence on a 1:1 mapping
between heritable variation and mutation.I was jus pointing out that the concept of heritable variation
has a somewhat different origin, and that in essence it was
not the same as a mutation.

quote:

---

You want to define mutation to be heritable variation, even
though it's not, but I'm not sure why.

​

This has never been my position. This is in fact a blatant misrepresentation of my position.

---

You said there IS a 1:1 mapping between mutation and heritable
variation -- what was I supposed to think that meant?

quote:

---

As to my definition being conditional, it isn't. It is you that wants to add all these conditional terms and caveats such as predictable and that it only applies in a germ cell line. The only thing I am saying is that it is a heritable change even if it is not likely to be inherited due to being in a somatic cell which is normally not cycling. By your criterion a mutation in the germ line of an animal wouldn't be a mutation as long as that animal failed to reproduce.

---

I've not said that it's only a mutation if it's in the germ-line
I said it's not heritable variation unless it's in the germ-line.And the term is 'heritable' i.e. capable of being inherited,
that does not require it to BE inherited and I cannot see how
it could be interpreted as such.

quote:

---

As to your qualitative difference, it is no difference in the mutation which was my point, there is no way to distinguish between the 2, its provenace is only a qaulitative difference in the causation, not in the mutation.

​

I see absolutely no requirement for the source of a mutation to be natural? Why on earth should there be one?

---

I see a need to inicate the difference, for various reasons(one
of which has to do with intelligent intervention). I accept that
it does not need to be part of the basic defintion though based
upon your reasoning above.

quote:

---

You seem to want a more and more restrictive definition of a mutation; it has to be unpredictable, it has to have a natural origin, it has to be in a germ line cell none of these are inherent in your initial definition of a mutation as a 'copy error' where in fact you only link that to the concept of heritable variation.

---

I still view it that for it to be a mutation it is inherently
unpredicted, but I hace never said that it isn't a mutation
if it's in a somatic cell. That confusion can only arise if
one requires a 1:1 mapping between heritable variation and
mutation (which I do not see to exist).

quote:

---

Well, surely a mutation is just an uncorrected DNA
copy error introduced during cell-division.

​

Just because some of an organism's cells have suffered
such a copy-error doesn't mean that that will be passed on
to it's offspring ... but if it does it becomes a
heritable variation.

---

The above covers my opinion (removing 'DNA') so that we don't
cycle back through epigenetics.I still think, though, that predicted changes could be imposed
by a cellular mechanism -- that eventuality makes 'unpredictable'
a necessary constraint on what is or is not a mutation.Even if no such mechanisms are uncovered it remains a logical
possibility.

I have more-or-less conceded this point and dropped it.The Norwegian deer have suffered an induced, potentially heritable,
but entirely unpredictable set of changes.I've not said anywhere that somatic changes are not mutations
only that not all mutations represent heritable variation.

quote:

---

Can you give me a natural example of an "anticipated" mutation?

---

The way I view it the above is contradictory That's why I say a mutation is an unpredicted change.

quote:

---

How can you tell the difference between a deliberate mutation and an accidental one (excluding site directed mutagenesis).

---

After the fact, I would expect that you cannot (unless there remain
some chemical markers of some sort).If I build a hill I cannot claim that it was caused by geological
processes (even if I used some geo-mechanism-inducing ray).The end result does not define the causative act.That said, I have agreed to drop that aspect (but would prefer
some comment on it -- like natural Vs. artificial is used in
reference to selection).

I did notever claim any such thing, if I said it then as I asked before show me where, you bought the term variation in at post 42. Since I had defined a mutation as a heritable change I should think it would be expected that I would claim a 1:1 mapping between mutation and heritable change. But as I have pointed out a change is not neccessarily a phenotypic variation, which is what you seem to be talking about. It arguably is a variation, but not one reflected in the phenotype, which is what you seem to be trying to claim I said. Well neither can I, but it is what you suggest by claiming that you can't consider somatic mutations to be heritable at all. They are not heritable in the organismal lineage but they are still heritable.The entire point of this thread was to produce a clear and versatile definiton of mutation.We seem to be going round in circles now, and we are right back at the point where we have to ask how unpredictable your mechanism has to be, is the VDJ system unpredictable enough?Your most frequent argument seems to be based on a claim which I never made, i.e. a 1:1 mapping of mutation and heritable phenotypic variation.

unpredicted change is not a great way of describing it...I can predict i.e. calculate the mutation rate for a given sequence replicated by specific polymerases based on empirical study of their error frequency. I can be almost certain that mutations will occur in almost all replication cycles..thus mutation per se is not unpredicted..it is expected. Which base change occurs is not equi-probable (to steal a line from Taz)Building a hill is not the same as using a mutagen on mice. You set out with the goal of building a hill which is a specific physical construct. ENS mutation of mice is blasting them with a mutagen and then seeing what mutations (or phenotypes) occur in any surviving progeny of the mutagenized mice...you can screen for a specific type of phenotype after the fact...but you cannot a priori go out and say I am going to make a mouse that runs around in circles like the twirler mutant mice, hit them with the mutagen and expect that result.I also take issue with the idea that somehow if you control a natural process to induce change versus a natural process occuring at random inducing a change that the mechanisms are entirely different....other than the selective force being the taste of the dog breeders, what is the biological difference between selection of dog breeds and natural selection? Similarly, what is the chemical difference between a mutation induced by me shining uv light on my arm and a mutation I get from standing out in the sun too long?

If you cannot say which base-change will occur then you
haven't predicted THE mutation you have predicted that a
mutation will occur.If I roll a die and say it will fall on a side I haven't
made a prediction.Not sure how gene therapy techniques fits with my views though,
although I have read that these don't always have the desired
effect (read something about a strange effect from a modified
gene getting inserted 'too close' to something else and
causing problems).I have conceded the 'causation' clause though.I still think it is relevant in some respects to know
if the mutation was intentionally provoked, but agree that
it is not relevant to a defintion of 'mutation'.

I'll also drop any reference to phenotypic variation
since I was neglecting synonymous changes.The 1:1 thing is a problem depending on what you consider
to be a relevant definition of 'heritable'.I was thinking of heritable in terms of 'passed to subsequent
instances of individuals within the species.'That seems the meaning relevant to evolution.

What definition would be relevant would depend what the context was. Obviously if you were talking about the evolution of a cancer, as this review does, then it would be entirely relevant to think in terms of heritability within a cell line. If you were discussing the evolution of pentadactyly in vertebrates then it wouldn't be.These are both examples of evolution however. What mutations are heritable will depend on the level you are looking at.

I cannot access the article.Is the cancer article using 'evolution' loosely (like
'development') or in the ToE sense of descent with modification
causing a change in allele frequency?Personally I feel that defining heritability in that way
obscures the issues, rather than helps to illuminate them.A somatic cell that is copied from an existing somatic cell can
be described as a descendent of that cell (in the same way that
all of our cells are 'descended' from the zygote). In some ways
I quite like that, especially since I have wondered whether it
would be valid to consider multi-cellular organisms as
ultra-dependent colonies.BUT it is not relevant to evolutionary theory (which is the
study of the origins of diversity of species). Whichever way
you cut it, somatic mutations do not enter subsequent generations.Heritable variation in a ToE sense does NOT include somatic
variability.A mutation is ANY change between the 'parent' cell and the
'offspring' cell -- I've even droppped unpredicted for the
moment even though I think it is a fundamental feature of
mutations.If one wishes to define mutation then 'change' or 'copy error'
covers it, but then we all knew that in the first place
didn't we?

That is why I said that building a hill is not a good comparison with the mechanism of selecting a phenotype...you will always be in the position of not knowing if the mutation you want will occur or not..you can just up the overall chances of mutation and screen for the result you want...the hill building analogy would only work if you said you want a hill of a specific shape and size so you bombard the earth with large meteors and screen for the hill you wanted to produce.rolling die also does not work...you know a side must come up...you don't know that a mutation has to occur.gene therapy is a bit of a nightmare I think. For one thing, your construct often lands somewhere you don't want it to and is shut down, it can recombine and produce crap, or since most biochemical pathways are not really understood, it can interact with endogenous cellular factors and produce unexpected results....I agree that in some cases you want to know the potential origin of certain classes of mutations...does living near a nuclear plant increase your chance of getting cancer and other diseases due to an increased mutation rate for example? But seen at a chemical level, it is not really different.