Poo Poo It or Something

[Glenn]

"The random occurrence of mutations with respect to their consequences is an axiom upon which much of biology and evolutionary theory rests1. This simple proposition has had profound effects on models of evolution developed since the modern synthesis, shaping how biologists have thought about and studied genetic diversity over the past century. "
...
"Our discovery yields a new account of the forces driving patterns of natural variation, challenging a long-standing paradigm regarding the randomness of mutation and inspiring future directions for theoretical and practical research on mutation in biology and evolution."

https://www.nature.com/articles/s41586-021-04269-6

[Öö Tiib]

Science has fair amount of garbage in it, just less than everything else.

In your cite they sequenced seedlings. Seeds that had mutations that made it difficult or
impossible of becoming seedlings were therefore filtered out by method of experiment.
Resulting sequences having entirely random distribution of mutations in those would be
rather unexpected with said method. But they somehow write that it was discovery
that the distribution of mutations measured that way wasn't entirely random.

Hopefully I was too harsh and did not grasp some nuance. But anyway was interesting to
read, thanks Glenn.

[jillery]

On Mon, 17 Jan 2022 17:41:16 -0800 (PST), Glenn <GlennS...@msn.com>
wrote:

What the cited experiment shows is that highly conserved genes, ex.
cytoplasmic translation, are less likely to mutate than specialized
genes, ex. response to chitin. It documents that highly conserved
genes are more likely to be associated with various histones that
stabilize DNA and promote their repair.

The above experiment isn't a refutation of your first quote, nor does
it challenge that paradigm. It's well-documented that different kinds
of mutations have different mutation rates. What the cited experiment
shows is that different kinds of genes also have different mutation
rates. Regardless of the rate, the mutations themselves still aren't
correlated with the fitness of organisms to their environment. IOW
mutations are still random with respect to their consequences.

This is consistent with an evolutionary model. Organisms which
preserve critical genes are more likely to reproduce and pass on their
that trait, while lineages which preserve genes that no longer fit the
environment are less likely to reproduce and pass on that trait.

--
You're entitled to your own opinions.
You're not entitled to your own facts.

[Lawyer Daggett]

Never trust a paper that claims to challenge a long-standing paradigm.
If you see one, it is essentially a given that they will manufacture this
straw-man paradigm by pressing gross over-simplifications of what was
already known and over-stating their contributions to new insights.
(see just about every paper that claims to challenge the Central Dogma)

The cited paper should have been rejected without a major rewrite.
The reviewers should have required the authors to avoid the equivocation between
mutations that are observed to survive over time and mutations occurring in the first
place. It isn't that the authors appear confused by the distinction, however, their
language invites confusion, again and again and again. A naive reader will be led astray.

They are also guilty of an all too common sin, of playing up their work as being
somehow revolutionary. It isn't in the manner that they imply. See where they
write "In conclusion, we uncovered associations between mutation frequencies
and biochemical features known to affect DNA repair and vulnerability to damage."
Of course if you read up to that point, you'll notice they've referenced publications
that have already discovered those very things. In fact, it's a rather heavily researched
area that's been beaten on pretty hard. They aren't adding any new mechanisms.
They aren't even adding any new broad observations, just some more specifics.
The specifics are valuable, and in a sense the use of the specifics to illustrate
aspects of non-random mutagenesis offers up an opportunity for a useful didactic,
but the writing is over-the-top.

[broger...@gmail.com]

So, so true. It's just a minor annoyance if you have the background to read the technical parts of the paper and, especially the figures. But if you are, for example, a creationist without much background in the details of the science and you think you really ought to read the original papers about whatever it is that evolutionnews has you excited about, then, when you read the paper, the only parts you can understand are exactly the exaggerated prose bits, the "we know nothing about X" sentence that begins the paragraph in which the details of all the thing we know about X are laid out, or the "we propose a new paradigm for Y" sentence that begins a paragraph in which a minor technical adjustment to the paradigm for Y is suggested.

I've forgotten his name, but there was a creationist posting here a few years back who decided he would learn science by reading papers, explicitly refusing to read any background texts first. So he'd come here armed with just the sorts of sentences you describe, plucked out of the "overselling" parts of the paper, with no idea what the actual experiments were or what they meant, and say "Look, right here in this peer-reviewed paper it says we don't have any understanding of X," and then work to the conclusion that a designer is required to explain X.

I don't think it's possible to write papers in a way that forestalls quotemining or deliberate misunderstanding by creationists, but I'd certainly be in favor of toning down the sort of hype you describe.

[Ernest Major]

They recognise your point, but consider it a minor distortion.

"very few mutations, only those causing inviability or sterility, should
have been removed by selection"

Any coverage of inviability and sterility in the work they cite for this
isn't immediately obvious.

>
> Hopefully I was too harsh and did not grasp some nuance. But anyway was interesting to
> read, thanks Glenn.
>

I would have restated the original principle as "for each mechanism -
which is a combination of damage and repair mechanisms - of mutation,
the probability of a mutation occurring is independent of its associated
selection co-efficient, but as the spectrum of selection coefficients
differs between mechanisms, given a sufficiently large population we can
expect that mechanisms with a spectrum biased towards negative selection
coeffecients - e.g. frame shift mutations compared to point mutations -
will be preferentially suppressed". That prediction of mine is
essentially the result that they are claiming, though they are claiming
a finer-grained discrimination than I would have naively expected.

Whether a sufficiently large population was achievable in nature was a
question to which I did not have a means of answering. This paper seems
to have gone some way towards answering it.

--
alias Ernest Major

[RonO]

How does this in any way help IDiocy/creationism? How long have we
known about mutation rate bias between different sequences? This still
has nothing to do with changing anything about whether specific
mutations happen when needed. Just recall the old bacterial work.
Under stress the SOS pathway is triggered and more mutations happen and
more mutations that are selectable under those stress conditions that
triggered the SOS are produced and selected for. The problem is that
mutation rate is increased across the genome and not just in the genes
needed to cope with the current stress.

All these guys are claiming is that mutations occur in different
sequences at different rates. They are still arbitrary compared to what
the organism is dealing with. There is no forethought or design
involved. This is just what happens, and the mutations occur and are
selected for or against or not under any noticeable selection. So what
does this do for IDiocy?

The authors are even claiming that a lot of the mutations that they have
observed do not show evidence of selection. What they are calling
adaptive mutation bias is the lower frequency of mutations where they
they would probably do the most damage. The adaptation was to reduce
the frequency of mutations in essential genes and gene regions. Bad
mutations happen all the time, but they would happen less often in
Arabadopsis due to epigenetic markers associated with decreasing the
rate of certain types of mutations.

Ron Okimoto

[Burkhard]

Lawyer Daggett wrote:
> On Monday, January 17, 2022 at 8:45:33 PM UTC-5, Glenn wrote:
>> "The random occurrence of mutations with respect to their consequences is an axiom upon which much of biology and evolutionary theory rests1. This simple proposition has had profound effects on models of evolution developed since the modern synthesis, shaping how biologists have thought about and studied genetic diversity over the past century. "
>> ...
>> "Our discovery yields a new account of the forces driving patterns of natural variation, challenging a long-standing paradigm regarding the randomness of mutation and inspiring future directions for theoretical and practical research on mutation in biology and evolution."
>>
>> https://www.nature.com/articles/s41586-021-04269-6
>
> Never trust a paper that claims to challenge a long-standing paradigm.

Yep. I've been extremely tempted to desk-reject any paper that makes
that claim with this boilerplate:

"Your paper is paradigm shifting. That means as a member of the old
paradigm, I cannot possible understand or evaluate it, by definition. It
is after all incommensurable to what has been done before, and therefore
nobody can say if it is better or worse. As per Feyerabend, I can only
recommend that you set up your own journal where people within your
paradigm publish, while you wait for my generation to die. Yrs..."

[Glenn]

All excellent responses so far!

[J. J. Lodder]

Burkhard <b.sc...@ed.ac.uk> wrote:

> Lawyer Daggett wrote:
> > On Monday, January 17, 2022 at 8:45:33 PM UTC-5, Glenn wrote:

[-]

> >> ... "Our discovery yields a new account of the forces driving patterns
> >> of natural variation, challenging a long-standing paradigm regarding
> >> the randomness of mutation and inspiring future directions for
> >> theoretical and practical research on mutation in biology and
> >> evolution."
> >>
> >> https://www.nature.com/articles/s41586-021-04269-6
> >
> > Never trust a paper that claims to challenge a long-standing paradigm.
>
> Yep. I've been extremely tempted to desk-reject any paper that makes
> that claim with this boilerplate:
>
> "Your paper is paradigm shifting. That means as a member of the old
> paradigm, I cannot possible understand or evaluate it, by definition. It
> is after all incommensurable to what has been done before, and therefore
> nobody can say if it is better or worse. As per Feyerabend, I can only
> recommend that you set up your own journal where people within your
> paradigm publish, while you wait for my generation to die. Yrs..."

Bravo!
But your recipe is just what happens regularly,
with controversial subjects.
(like quantum conciousness for example)

People engaged in it set up their own channels,
and referee each other, and it can go on forever.
Fine with Feyerabend, but Lakatos would have said 'degenerating!',
after a sufficiently long time. [1]

The only way to get rid of it is to stop funding it,

Jan

[1] I have heard Feyerabend speaking, IRL,
and he made a lot more sense than your caricature of him.
But he couldn't help being the eternal provocateur.

[Athel Cornish-Bowden]

On 2022-01-18 09:43:13 +0000, Lawyer Daggett said:

> On Monday, January 17, 2022 at 8:45:33 PM UTC-5, Glenn wrote:
>> "The random occurrence of mutations with respect to their consequences
>> is an axiom upon which much of biology and evolutionary theory rests1.
>> This simple proposition has had profound effects on models of evolution
>> developed since the modern synthesis, shaping how biologists have
>> thought about and studied genetic diversity over the past century. ">
>> ...> "Our discovery yields a new account of the forces driving patterns
>> of natural variation, challenging a long-standing paradigm regarding
>> the randomness of mutation and inspiring future directions for
>> theoretical and practical research on mutation in biology and
>> evolution.">> https://www.nature.com/articles/s41586-021-04269-6
>
> Never trust a paper that claims to challenge a long-standing paradigm.

Never take seriously someone who claims to be a scientist and talks
about paradigms at all. OK, that's a bit of an exaggeration, but
"paradigm" is a word much loved by journalists and authors of popuar
articles, but rarely used in any real science publications.

What emerges from Thomas Kuhn's book is a man deeply in love with his
own ideas. If it were not for the fact that not appearing ridiculous
was clearly not very high on his list of priorities one might feel at
times that he was trying to see how many “paradigms” he can pack into
one paragraph. A reviewer said that there were 677 uses of the word in
the book. I haven’t counted, and my first reaction was that that was an
exaggeration, but at two or three on every page it’s not impossible.

> If you see one, it is essentially a given that they will manufacture this
> straw-man paradigm by pressing gross over-simplifications of what was
> already known and over-stating their contributions to new insights.(see
> just about every paper that claims to challenge the Central Dogma)

The problem with those papers is that their authors not only
misunderstood what the central dogma actually is, but they didn't
bother to read Crick's subsequent explanation.

>
> The cited paper should have been rejected without a major rewrite.The
> reviewers should have required the authors to avoid the equivocation

> betweenmutations that are observed to survive over time and mutations

> occurring in the first
> place. It isn't that the authors appear confused by the distinction,
> however, their
> language invites confusion, again and again and again. A naive reader
> will be led astray.
> They are also guilty of an all too common sin, of playing up their work
> as being
> somehow revolutionary. It isn't in the manner that they imply. See where they
> write "In conclusion, we uncovered associations between mutation

> frequenciesand biochemical features known to affect DNA repair and

> vulnerability to damage."
> Of course if you read up to that point, you'll notice they've
> referenced publications
> that have already discovered those very things. In fact, it's a rather
> heavily researched
> area that's been beaten on pretty hard. They aren't adding any new mechanisms.
> They aren't even adding any new broad observations, just some more

> specifics.The specifics are valuable, and in a sense the use of the
> specifics to illustrateaspects of non-random mutagenesis offers up an

> opportunity for a useful didactic,
> but the writing is over-the-top.

--
Athel -- French and British, living mainly in England until 1987.