Is Random Mutation Totally "Random" In Biology/Evolution?

[kyphysics]

When I see the term random mutation in popular writings on evolution (social sciences major here, so please forgive my ignorance), I wonder what it precisely refers to.

I understand we can have have gene mutations due to exogenous factors, such as exposure to UV light, or from "errors" in the DNA copying process. Are there any other causes of mutations?

However, I'm wondering if by "random," scientists mean there is no "logical cause" of these mutations. Clearly, UV radiation would be a cause. And then when an error is made in DNA copying, can we not locate a cause for the error? I guess I'm not sure why we call it "random mutation" and not some other qualified mutation? And is there ever truly a kind of totally random - like a spontaneous, cause-less - mutation?

Does this question make sense?

 

[jim mcnamara]

Random means it cannot be predicted successfully - for one interpretation.

Make sense - sort of

This is a massive topic
Bottom line - without mutations most populations could never adapt to long term changes, and would go extinct.

Some causes of genetic changes (All simplified):

Genetic drift as meiotic drive caused by meiotic failure -
Genes can get changed when chromosomes do not crossover "correctly" during gamete formation or mitosis. Can result in scrambled DNA. Cultured organisms that start with identical DNA, become changed over multiple generations. They are no longer all identical. Most of the mutations are detrimental, at least short term. So they may persist as so-called recessive alleles.

The mutation rate you are thinking of relates to chemical (think diet for animals and heavy metal deposits for plants) and radiological (environmental) damage to DNA. Most of these are detrimental, too.

Overall:
The level of mutations in a population is beneficial to maintaining that population over time.

Even though it will result in some failures to reproduce. Successful parents have to shuffle the deck of DNA cards dealt to offspring -- Why?
changing climate or water availability, or geological processes:
Isolation from food supplies or soil types they are adapted to.
New diseases or vectors.
New species migrating into the area - resource competition
Extinction of support species - symbiosis. Maybe a pollinator for plants.

 

[BillTre]

The phrase "Random Mutation" does not entirely project a clear concept.
There are many kinds of mutations caused by a variety of molecular/cellular mechanisms.

Location of mutations: The positions of mutations within a genome can appear random at some level of resolution.
I've been involved in a variety of mutageneses using a variety of methods (X-rays, UV, P-elements, ENU (vertebrate chemical mutagen), Psoralen (a chemical that can cross links DNA it they are together exposed to UV). They each have different advantages and limitations, including various ways of not making random mutations.
The most often ideal goal of a mutagenesis is (starting from a known isogenic (all individuals in the population are genetically identical)) to generate mutations at random throughout the genome. This distribution of new mutations that are generated through what ever skewed distribution in the genome. The rate new mutations are generated at, is controlled, such that number of mutations in a single germline cell, doesn't generate more mutations, per germline cell, can be identified through genetic breeding trips and screening/selections.
The ideal desired use of a good mutatgeneises is to do a saturation mutatgenesis. In a saturation mutagenesis, you collect so many mutations that by reasonable chance, you have collect mutations in every gene that your screen (looking for mutants) can capture phenotypes of. The mutations that are found, define pathways by the genes, identified by mutations, that the mutagenesis finds.

This provides me with a complex view of random mutations.
Realize that a real world (or natural) rate of spontaneous mutation will be a summation of many different processes going on, in the uncontrolled "normal" or "real" world.

Overall Rate:
In a lab, things can be separated and controlled. But that doesn't happen in the "wild". In some places, at some times, different mutagenic rates will vary for the different mutation generating mechanisms.
In the sense that its an uncontrolled and unanalyzed mess, the resulting total rates can be considered, first look random.
Chemical and radiation mutagenic exposures can easily be controlled in the lab.
In nature, they could vary widely from place to place, and from time to time.

Location is not always random:
Some mutational processes target or avoid certain genetic sites. There are lots of plausible proposed reasons why, but I don't know of clear demonstrations of most of them.

Biological Mutagenesis, like the transposition of P-elements within a genome, will often favor particular genomic sites (defined by the sequence they insert into). This is presumably due to particular molecular cellular reasons limiting or block access to particular genomic sites. This skewing of mutation by this particular method is often quite local, which might leave other regions of a particular gene available for attack. P-element use is well controlled by researchers to mutate and ID genes. Although, it might not produce a pure random collection of mutations, it provides a lot and the mutations are easily tied to the sequence they are affecting.

Chemical and radiation caused mutations are considered to have a more random distribution of the sequences they affect (nothing should get in the way of X-rays, many mutagenic molecules are not large (like a protein) and therefore should not be so easily blocked).
(However, they do not have a useful marker linked to the mutated sequence, so they require more work to link the genetic mutation (determined by genetic patterns of inheritance) to a particular molecular sequence in some region of the genome (which is an important component of molecular analysis).)

Types of Mutations Produced:
Different causes of mutations make different kinds of mutations. There is a specific distribution to the innds of mutations each method will produce: point mutations, break point mutations, short deletions, sequence insertions, a mixture of each.

So the amount, distribution through the genome, and the kinds of mutations could vary uncontrollably in a non-lab situation.
It might look random mutations, but there would probably underlying non-randomnesses in its distribution.

 

[jim mcnamara]

@BillTre did a nice, more detailed answer. If the terminology in either answer is too much please let us know.

My favorite positional mutation is called 'jumping genes' aka transposons. Students always liked the concept.
PS: About 40% (or more)of the changes in the order of DNA segments can be considered an active mutation, versus 'junk DNA' - exons versus introns.
See this article in Nature:
https://www.nature.com/scitable/topicpage/transposons-or-jumping-genes-not-junk-dna-1211/

 

[Ygggdrasil]

When I see the term random mutation in popular writings on evolution (social sciences major here, so please forgive my ignorance), I wonder what it precisely refers to.

I understand we can have have gene mutations due to exogenous factors, such as exposure to UV light, or from "errors" in the DNA copying process. Are there any other causes of mutations?

However, I'm wondering if by "random," scientists mean there is no "logical cause" of these mutations. Clearly, UV radiation would be a cause. And then when an error is made in DNA copying, can we not locate a cause for the error? I guess I'm not sure why we call it "random mutation" and not some other qualified mutation? And is there ever truly a kind of totally random - like a spontaneous, cause-less - mutation?

Does this question make sense?

You are right to identify that there are mutations due to exogenous factors (e.g. UV light or mutagenic chemicals), but there is also intrinsic error rate for the enzymes that copy DNA. These intrinsic errors come from the polymerase enzymes that copy DNA incorrectly matching nucleotide bases during the copying process, resulting in a mutation. There is a "logical cause" of the mutation (error from the DNA copying process), but these mutations do not result from exposure to any extrinsic factors. The intrinsic error rates of the polymerase enzymes can be important for generation mutations for evolution (e.g. many viruses have polymerase enzymes with somewhat high error rates, which generates diversity as the viruses replicate, giving a large pool of mutations for natural selection to work with, explaining why many viruses can evolve quite quickly).

Over the past few years, there has been some debate in the scientific community as to the relative contributions of intrinsic vs extrinsic mutations in the development of cancer. This is an important question as it relates to what fraction of cancers might be preventable vs what fraction arise unavoidably due to the "bad luck" of which errors pop up as cells copy their DNA. Here are some previous physics forums threads on the topic:
https://www.physicsforums.com/threads/bad-luck-is-primary-cause-of-most-cancer.790125/
https://www.physicsforums.com/threads/what-causes-cancer-bad-luck-or-bad-lifestyles-comments.848662/
https://www.physicsforums.com/threads/cancer-causes-circa-67-are-replication-error.908823/

 

[jim mcnamara]

We also did not mention epigenetics - chemical changes to DNA that are associated with environmental and developmental changes. They can "slow down" or "turn off" a particular gene for example. In general, these changes are not passed on to offspring.

Example: Twin studies show that identical twins (they start out with as close to the same DNA as we can get for human siblings)- one a smoker, one not a smoker, have radically different epigenetic changes.

see:
https://www.nature.com/articles/pr2007129

abstract
Acute environmental factors are directly associated with epigenetic-dependent disease phenotypes, as demonstrated by the increased CpG-island promoter hypermethylation of tumor suppressor genes in the normal oral mucosa of smokers.

In plain English, hypermethylation is lots of chemical groups parked onto the DNA of genes that help to stop oral tumors. It turns them off so they do not work anymore.

 

[atyy]

However, I'm wondering if by "random," scientists mean there is no "logical cause" of these mutations. Clearly, UV radiation would be a cause. And then when an error is made in DNA copying, can we not locate a cause for the error? I guess I'm not sure why we call it "random mutation" and not some other qualified mutation? And is there ever truly a kind of totally random - like a spontaneous, cause-less - mutation?

No, of course not. "Random" in the context of biology almost always means resulting deterministically from many causes of which we do not have detailed knowledge. It is the same notion of randomness in classical statistical physics, where the notion of random motion of gas molecules does not mean that Newton's deterministic laws do not apply.

 

[jim mcnamara]

I think that random is hard to grasp the way it is used by science folks.

@atyy your explanation is clear and correct. I like it. However it may not do much to clear the OP's confusion.
Why? Because you used several terms that are "local" to science.

For example:
My take is that the a priori view of random used outside science goes something like this:
You are hiking out on a remote desert trail. Out of nowhere, a brick falls out of the sky and really hurts you.

Helen Quinn wrote a Science (AAAS) opinion piece on why scientists have problems with certain words when talking to non-science folks. Random is definitely on that list, IMO. Right along with 'theory', 'I believe', 'coincidence' ...

Her opinion:
https://ned.ipac.caltech.edu/level5/March07/Quinn/Quinn.html

This is an area where popular science writers can and sometimes do confuse readers by making things 'too simple'. So, overall it is hard to do well.

 

[atyy]

Her opinion:
https://ned.ipac.caltech.edu/level5/March07/Quinn/Quinn.html

The one I really hate is "work" unless I'm the boss

 

[kyphysics]

*reading through and trying to digest everyone's answers before replying*

 

[Laroxe]

While changes can be described as random there are lots of influences that act on the process and many of these are under the influence of other genes. The whole process of cell division is subject to controls on the copying of the genes their separation, the cell separations and the end result of each stage subject to checks. Its a complex process and mistakes are not uncommon, the great majority of this will label the cell for destruction.

There is the potential for a mistake at every division so considerable resources are put into the checking mechanisms but even then some mutations will slip through the net these will usually have little functional significance. The number of cell divisions will impact on the overall number of these altered genes so by the time humans reach fertility they will have a large pool of altered genes which provide a population which natural selection might draw upon.

At the end of the fertile period it seems biology rather looses interest in investing resources into the quality control mechanisms of cellular reproduction, so more of the mistakes survive in our genome and accumulate. This has serious health implications but without reproduction none of these changes can be transmitted and so have no evolutionary significance. It might be suggested that the possibility of mistakes is an essential part of the whole system.

So while random may not be the best word, its hard to think of another, we don't know why some occur, why some survive, and we can't really predict the outcome. Its one of those areas that explain why the Greeks invented the idea of the Fates.

 

[kyphysics]

No, of course not. "Random" in the context of biology almost always means resulting deterministically from many causes of which we do not have detailed knowledge. It is the same notion of randomness in classical statistical physics, where the notion of random motion of gas molecules does not mean that Newton's deterministic laws do not apply.

So, I think this mostly answers my question.

I was caught up in the semantics of it all and wondering why the qualifier "random" was added to the word mutation. Why not just say "mutation," instead of "random mutation" when talking about evolution in biology? *serious question*

Yeah, I figured that scientists probably don't REALLY mean there is complete randomness to these actions - just that the causes may not be readily known/available at the time of the event (or discovery of it).

Maybe my entire confusion was just one of word choice and semantics.

 

[atyy]

I was caught up in the semantics of it all and wondering why the qualifier "random" was added to the word mutation. Why not just say "mutation," instead of "random mutation" when talking about evolution in biology? *serious question*

Historically, it is because the "randomness" of the mutations was contrasted with the "non-randomness" of natural selection.
https://evolution.berkeley.edu/evolibrary/article/evo_32

Of course, technically, in mathematics, determinism is a specific sort of randomness. So the question is not so much whether something is random or not, but what the probability distribution is - is it uniform, Dirac delta measure, Gaussian, Poisson, etc.

We do have useful mathematical models of evolutionary and ecological phenomena, and these use "random" in a specific mathematical sense. For example, the Hardy-Weinberg equilibrium describes what the stable allele frequency is when mating is "random", and there are no mutations (among other assumptions): https://en.wikipedia.org/wiki/Hardy–Weinberg_principle.

You can read about the quite interesting history of attempts to mathematically model evolution (still many ongoing debates, see the entertaining debate on "eusociality" in the links below).
https://en.wikipedia.org/wiki/Fisher's_fundamental_theorem_of_natural_selection
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3279739/
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3836173/

 

[Eagle9]

jim mcnamara
BillTre
Ygggdrasil
Hello people
I think that this thread is suitable place to ask my question and not to open a new thread:

Now I read the book “Evolving Brains” by John Morgan Allman. The book is interesting in many aspects and particularly it says many things about linkage between gene duplication and evolution, for example:

The different serotonin receptors were created by a series of gene duplications. (I will have much more to say about the role of gene duplication as an evolutionary mechanism in Chapter 3.)

In 1918, Bridges reported repeated sets of banding patterns in the chromosome map, which presumably contained repeated sets of genes. He proposed that the duplications offered "a method for evolutionary increase in the lengths of chromosomes with identical genes which could subsequently mutate separately and diversify their effects."

Homeobox sequences were soon found in genes throughout the animal kingdom: in hydra, planaria, sea urchins, nematodes, beetles, locusts, amphioxus, fish, frogs, chickens, mice, and humans. These genes were all products of gene duplications at different times in the evolutionary past and were derived ultimately from a primordial gene in the common ancestor of all these animals that contained the homeobox DNA sequence.

Gene duplications provide the raw material for evolutionary change. However, many mutations of duplicate genes have negative consequences for the organism, such as the mutation of Emx-1, which results in the failure of the corpus callosum to develop, and the mutation of Emx-2, which produces split brains.

But this book says nothing how this gene duplication(s) occur on molecular level. If I guess correctly gene duplication is essentially the same as transposable elements (transposons)? Or perhaps gene duplication is something different?

 

[Ygggdrasil]

But this book says nothing how this gene duplication(s) occur on molecular level. If I guess correctly gene duplication is essentially the same as transposable elements (transposons)? Or perhaps gene duplication is something different?

Transposable elements are one source of gene duplication, but there are others involving errors in DNA replication or recombination:
https://bio.libretexts.org/Bookshel...nomes/18.4D:_Gene_Duplications_and_Divergence

 

[BillTre]

In addition, to the sources of duplicate genes @Ygggdrasil mentioned,
Whole Genome Duplications have occurred intermittently several times in evolution.
These provide lots of new material for possible evolution at a discrete time in evolution.
The sub-functionalization mechanism mentioned in @Ygggdrasil's link, it then thought to sort through these duplicated genes for potentitally useful ones in the generations following a big duplication.
These new genes could rapidly (on an evolutionary time scale) be assembled into new processes, using genes with divergent structures and/or expression patterns.

Among vertebrates, evidence of two genome doublings have been found early in vertebrate history
(after amphioxus, before sharks), while another duplication has occurred in the lineage leading to many fish. Some other lineages (like those leading to sturgeons) have evidence of even more duplications.

 

[Eagle9]

Ygggdrasil
BillTre
Thanks guys

 

[jim mcnamara]

Polyploidy is an extreme example of whole genome partial duplication. Fragaria spp. hybrids (commercial strawberries) are polyploids, most of the ones we eat are 8n (8 sets of chromosomes instead of 2n or two).
Modern wheat is hexaploid, 6n, and semolina (Durum wheat) that is used for pasta is 4n.

Some of the food plants and almost all ferns are polyploids that occurred naturally. Strawberry and semolina wheat are thought to be a combination effort: some human, some natural.

"Ancient grains" is a collective term in the popular press for food items like farro (spelt, emmer, and einkorn are the names of the kinds of wheat) and semolina.

Polyploidy:
https://en.wikipedia.org/wiki/Polyploidy

Example discussion:
https://en.wikipedia.org/wiki/Strawberry

 

[Eagle9]

jim mcnamara
Thanks