Random Mutations and Biological Evolution

[Jay Kulsh]

Genes can mutate in about 7 ways, namely

• Missense mutation
• Nonsense mutation
• Insertion
• Deletion
• Duplication
• Frameshift mutation
• Repeat expansion

In each of these mutation types, there are multiple possibilities, for example a deletion can happen in any portion of the gene and it can be small or large.

All in all, there are at least hundreds of thousands of ways in which an average gene (that can build protein of 200 - 300 amino acids) can mutate. Only a very few of these mutation would be beneficial while most of them will be harmful.

To think that random genetic mutations kept leading to beneficial changes over hundred of millions years is like saying a person can keep winning lottery over and over, even though each time his/her odds are 1 in 100,000 or 1 in 1,000,000.

Is this line of thinking wrong? Could sheer randomness have led to so many refinements to various forms of life on the planet? Thanks.

[joigus]

Consider a frame-shift mutation. This kind of mutation is surely always deleterious. The individual dies even before it's born. So it's not like a person winning the lottery. Come to think of it, genetic material is some kind of template that uses the individual as a clever editing technique. I'm sorry if it sounds bleak, but that's the way it seems to be.

But, OTOH too, not all mutations are deleterious.

There are several molecular mechanisms of hedging you bets, or trying to have your cake and eat it too. I would say that anywhere along the molecular mechanisms of replication, transcription, and translation; whenever you see redundancy, there is at least the possibility of hedging your bets.

If you have a series of codons that give rise to synthesis of, eg, an essential enzyme; and you have this sequence repeated over and over, consider the possibilities: you will always have the essential enzyme synthesized, buy you will have copies of it, like when you have copies of a text from a previous template, so that you can work on them to produce several alternative versions, but never lose your backup.

Another mechanism of trying combinatorics is eukaryotic splicing, in which the organism tries several alternative "cut and paste" possibilities in between transcription and translation.

Paraphrasing Sir Humphrey Appleby in Yes, Minister: "Really, minister, the possibilities are endless" (when you have huge redundancy in your nucleic acids)

[swansont]

9 hours ago, Jay Kulsh said:

 Is this line of thinking wrong?

Yes.

[joigus]

Oh, sorry. I forgot:

9 hours ago, Jay Kulsh said:

Is this line of thinking wrong?

Yes.

[Bufofrog]

11 hours ago, Jay Kulsh said:

is like saying a person can keep winning lottery over and over, even though each time his/her odds are 1 in 100,000 or 1 in 1,000,000.

You mean like this?  Winners of more than one lottery.

11 hours ago, Jay Kulsh said:

All in all, there are at least hundreds of thousands of ways in which an average gene (that can build protein of 200 - 300 amino acids) can mutate. Only a very few of these mutation would be beneficial while most of them will be harmful.

I believe most mutations are neutral.  A beneficial mutation will be favored and will not be spread through the population by random chance, it will be spread because the mutation will tend to make those animals with the mutation more likely to reproduce.  So it is not like the lottery.

[Arete]

Most mutations are neutral. 

The lottery analogy fails due to the fact that evolution is a population level process. A diploid human genome experiences 175 mutations per generation on average. There were 3,745,540 human births in 2019 alone. That's 655,469,500 mutations across the human population in one year. Diploid human genome size is 6.4 Gb - so that population level mutational likelihood space is approximately 10% of the whole human genome in a single year. 

Next, "fitness" in evolutionary terms is discretely defined as genetic contribution to the subsequent generation. By definition, if a mutation is beneficial, it increases in frequency in subsequent generations (complications of neutral genetic drift aside). 

Therefore, "random" (they aren't actually random - only naïve with respect to fitness) explores significant portions of the human (or other species) total adaptive landscape every generation, and by definition, beneficial mutations, proliferate through the population throughout subsequent generations. 

[Jay Kulsh]

9 hours ago, joigus said:

veral molecular mechanisms of hedging you bets, or tryin

Arete,

Your reply seems more meaningful than others. Thanks for the link at "they aren't actually random". (Some others have simply stated that thinking of the original question is wrong, without saying a word why? I was seeking discussion, not a verdict.) Most random mutations may be neutral, but number of harmful such mutations would surely exceed beneficial ones.

Let me point out a scenario when beneficial mutation would have to occur in multiple places for evolution to move forward, increasing odds against it. For example, when oceanic creatures first started moving to land. To survive on land, they had to not be able to just drag/crawl their bodies in mud or dry land, but they also had to breath oxygen directly from air -- and face direct sunlight, etc. Therefore,  multiple random beneficial mutations had to occur in concert to be of any help in survival. Could a phenomenon like epigenetics have not played a role, where a mutation is not random but "effortful" or "interactive"? Thanks.

[Phi for All]

1 hour ago, Jay Kulsh said:

(Some others have simply stated that thinking of the original question is wrong, without saying a word why? I was seeking discussion, not a verdict.)

That's hardly fair. Nobody said thinking of the original question was wrong. And you ended the OP asking for a verdict:

16 hours ago, Jay Kulsh said:

Is this line of thinking wrong? Could sheer randomness have led to so many refinements to various forms of life on the planet? Thanks.

You asked if the line of thinking is wrong, and were given some great answers. Some of them you liked less than others.

[Arete]

25 minutes ago, Jay Kulsh said:

Most random mutations may be neutral, but number of harmful such mutations would surely exceed beneficial ones.

The fitness effects of a mutation are a moving target in a variable environment. While a mutation may be deleterious or neutral in one environment, it may be highly beneficial in another. E.g. antibiotic resistance in bacteria. Further, even if the net number of deleterious mutations outnumber the beneficial, the process of selection increases the likelihood of fixation of beneficial mutations. 

25 minutes ago, Jay Kulsh said:

Therefore,  multiple random beneficial mutations had to occur in concert to be of any help in survival. 

 This seems to be a version of the irreducible complexity fallacy - of which this is a good discussion of. As pointed out above, mutations do not need to be of benefit in the current environment to be prevalent or even fixed in an environment, so a trait can exist that has no contemporary function or benefit. Also, intermediate phenotypes are often more prevalent than many expect - as an example there are both air breathing fish and amphibious fish that can't breathe air that are extant today. 

Quote

Could a phenomenon like epigenetics have not played a role, where a mutation is not random but "effortful" or "interactive"?

Epigenetics and gene interactions may well be associated with beneficial mutation, however neither process could be described as effortful or interactive. 

Edited December 28, 2020 by Arete

[Jay Kulsh]

Arete,

I had taken the phrase "Interactive" from the title of your own citation: "Interaction-based evolution".

[joigus]

Just for the record, I agree of course that @Arete's answer was far better than mine in particular. I fail to see whether quoting me by an unintelligible fragment of text like,

13 hours ago, joigus said:

veral molecular mechanisms of hedging you bets, or tryin

was intended as criticism, or an accident when using the quoting function. Anyway, just in case it was addressed to me...

I also think the question was a very good one, and would like to see more questions like this, so that --I at least-- can learn more from the exchange. It's a good example of something everyone interested in biology have asked themselves at some point, and where simple "common sense" can, and does, mislead you.

As Phi said, you asked whether the line of reasoning was wrong. And although I'm no expert, I will insist here on what I think are the main flaws in my own words as a non-expert, and hoping either Arete or others can clarify further, or correct me where I'm wrong. Of the two main most common fallacies I see, one of them has already been pointed out. Namely: most mutations are neutral, not deleterious.

The other is the common misrepresentation of evolution that the "fulcrum" of it, so to speak, is the individual:

18 hours ago, Jay Kulsh said:

To think that random genetic mutations kept leading to beneficial changes over hundred of millions years is like saying a person can keep winning lottery over and over, even though each time his/her odds are 1 in 100,000 or 1 in 1,000,000.

(my emphasis) when the basic unit on which the lottery is played is the gene, or to be more precise, genotypic sequences. I think that's what Arete meant when they said,

5 hours ago, Arete said:

The lottery analogy fails due to the fact that evolution is a population level process.

[...]

Next, "fitness" in evolutionary terms is discretely defined as genetic contribution to the subsequent generation.

Beneficial changes accrue to the genetic sequence --not to the individual--, in the sense that it becomes more frequent in the population, even across species, as Arete has made transparently clear. I would say the genes for glicolysis have been impressively successful. They "couldn't care less" whether their host is a tomato or a dolphin, let alone this or that tomato or dolphin.

You can think of it in a rather cynical --but hopefully useful-- way, which is: The individual is just a carrier of the "meaningful" genetic sequence; once it's played its sexual, nurturing, etc. role, it's disposable. While the organism is alive, it benefits, of course, from using winner genes.

[Arete]

4 hours ago, Jay Kulsh said:

Arete,

I had taken the phrase "Interactive" from the title of your own citation: "Interaction-based evolution".

The full title is "Interaction-based evolution: how natural selection and nonrandom mutation work together" 

It clearly refers to the interaction of selection and mutation, and does not imply that an organism has some sort of ability to control the novel gametic mutations it inherits or passes on. 

[Jay Kulsh]

"some sort of ability to control the novel gametic mutations it inherits or passes on." I did imply that. Is this not what epigenetics is? There are certainly some examples where genes responds to environmental stresses and those modifications gets inherited. Epigenetics is in its infancy.

Till a few decades ago, ideas that are now accepted under epigenetics, were being laughed at. (Late Stephen J. Gould despised Lamarck with gusto. A day may come when Gould will be looked down upon for this and other reasons.)

Just look at Covid-19 virus. Within a year -- an extremely short time on the evolutionary scale -- it has mutated so that now it is more infectious. When 'random' mutations are favorable over and over -- despite having extreme odds against such occurrences -- then those mutations cannot truly be random. We just don't yet understand the underlying mechanism of the phenomenon.

[CharonY]

33 minutes ago, Jay Kulsh said:

Just look at Covid-19 virus. Within a year -- an extremely short time on the evolutionary scale -- it has mutated so that now it is more infectious.

You have to take into account that mutation rate and the equivalent of generation rate of viruses is massive even compared to a single-celled organism. Think about it that way, in humans a generation is about 20-30 years. In bacteria it can be as short as 30 minutes and virus particles can be produced in even a shorter time than that.

Viruses do not undergo cell division, but are rather produced by cells in copious amounts, of which there is plenty of opportunity to introduce errors. Then there is the fact that viral genomes are very condensed. I.e. there are not many intergenic regions, so mutations can often occur in coding regions (i.e. regions that encode RNA and proteins). So functional changes are simply more common and selection basically occurs on the level of interaction with their hosts (e.g. non-infectious mutations vanish and so might highly lethal ones).

In fact in a person with a long-term Covid-19 infection folks were able to isolate a variant that developed in the patient, which carried a few point mutations compared to the original strain with which the patient was infected with. 

In other words, when it comes to evolution, time scales are not absolute but depend on the organism (or virus in this case). 

[Arete]

2 hours ago, Jay Kulsh said:

"some sort of ability to control the novel gametic mutations it inherits or passes on." I did imply that. Is this not what epigenetics is? 

No, it isn't. The fact that environmental factors can create epigenetic mutations does not imply that the organism has any ability to control which mutations occur, nor what phenotypic changes they manifest. In fact, many are deleterious.

2 hours ago, Jay Kulsh said:

Till a few decades ago, ideas that are now accepted under epigenetics, were being laughed at. (Late Stephen J. Gould despised Lamarck with gusto. A day may come when Gould will be looked down upon for this and other reasons.)

Lamarckian evolution is not epigenetics, nor is Lamarckian evolution supported by epigenetic processes. 

2 hours ago, Jay Kulsh said:

Just look at Covid-19 virus. Within a year -- an extremely short time on the evolutionary scale -- it has mutated so that now it is more infectious. When 'random' mutations are favorable over and over -- despite having extreme odds against such occurrences -- then those mutations cannot truly be random. We just don't yet understand the underlying mechanism of the phenomenon.

Mutations are random with respect to fitness. Selection is the force that causes fixation of adaptive mutations. It is well understood. 

[Jay Kulsh]

On 12/29/2020 at 12:28 PM, CharonY said:

You have to take into account that mutation rate and the equivalent of generation rate of viruses is massive even compared to a single-celled organism.

 

CharonY,

Thanks for your valuable post. You are totally correct about how a 'beneficial' mutation in a virus would survive/thrive, even if most mutations are not 'beneficial' to its survival -- due to fast and furious way it multiplies.

But what about higher organisms like mammals where only a few progenies are produced in an entire year? Since most random mutations cannot be beneficial, how are they eliminated? They may not do any immediate harm, if genetic contribution of other parent mask their effect, but sheer accumulation of them in genome would be very bad in the long run. We do not even detect a whole lot of such harmful mutations. They certainly cause some genetic diseases but these are very rare -- afflicting probably less than 1% of the population.

 

On 12/29/2020 at 2:27 PM, Arete said:

Lamarckian evolution is not epigenetics, nor is Lamarckian evolution supported by epigenetic processes.

Search on Internet for "Epigenetics Lamarck" yields two kind of scientific discussions:

In the first group, Lamarck is remembered fondly while discussing epigenetics:
Epigenetic Change: Lamarck, Wake Up, You're Wanted in the ...
On epigenetics: we need both Darwin's and Lamarck's ... - Aeon

In the other group, introduction of name of Lamarck is not welcome:
Epigenetics is Normal Science, But Don't Call It Lamarckian ...
(PDF) Why epigenetics is not a vindication of Lamarckism ...

I obviously agree with the first group, while you favor the 2nd one. (I am not saying Epigenetics is equivalent to Lamarckism, only that it has shades of Lamarckism, and there is a lot more to come since epigenetics is its infancy.) Let us agree to disagree and I am closing this thread with you. Thanks for providing some good links.

[swansont]

2 hours ago, Jay Kulsh said:

CharonY,

Thanks for your valuable post. You are totally correct about how a 'beneficial' mutation in a virus would survive/thrive, even if most mutations are not 'beneficial' to its survival -- due to fast and furious way it multiplies.

But what about higher organisms like mammals where only a few progenies are produced in an entire year? Since most random mutations cannot be beneficial, how are they eliminated? They may not do any immediate harm, if genetic contribution of other parent mask their effect, but sheer accumulation of them in genome would be very bad in the long run.

Why? If the mutations are neutral, why would accumulating them be bad?

 

2 hours ago, Jay Kulsh said:

We do not even detect a whole lot of such harmful mutations. They certainly cause some genetic diseases but these are very rare -- afflicting probably less than 1% of the population.

Can you take the next step(s) in figuring this out?

[CharonY]

8 hours ago, Jay Kulsh said:

But what about higher organisms like mammals where only a few progenies are produced in an entire year? Since most random mutations cannot be beneficial, how are they eliminated?

They are not. Most mutations are neutral and not selected against. That is the definition of neutral. They keep them. 

Quote

but sheer accumulation of them in genome would be very bad in the long run

This is clearly not the case, a huge chunk of our DNA are areas that are duplicated, contain viral insertions and/or have otherwise accumulated bits and pieces. Typically there is no strong selection for eukaryotes to maintain a small genome size.  The eukaryotic amoeboid Polychaos dubium has a genome size of 670 Gigabases (compared to 2.9 of the human genome), for example. In prokaryotes there is some benefit of keeping genome sizes compact, in part because it allows faster replication time. 

[Jay Kulsh]

45 minutes ago, CharonY said:

They are not. Most mutations are neutral and not selected against. That is the definition of neutral. They keep them.

Most mutations may be neutral, but harmful mutations must still outnumber beneficial ones, if they are truly random.

[CharonY]

3 hours ago, Jay Kulsh said:

Most mutations may be neutral, but harmful mutations must still outnumber beneficial ones, if they are truly random.

It does not change anything about what has been discussed so far.

 

On 12/28/2020 at 2:32 PM, Arete said:

The fitness effects of a mutation are a moving target in a variable environment. While a mutation may be deleterious or neutral in one environment, it may be highly beneficial in another. E.g. antibiotic resistance in bacteria. Further, even if the net number of deleterious mutations outnumber the beneficial, the process of selection increases the likelihood of fixation of beneficial mutations. 

 

[Jay Kulsh]

You cite "Even if the net number of deleterious mutations outnumber the beneficial, the process of selection increases the likelihood of fixation of beneficial mutations."

It seems to me, this "process of selection" rationale works only for lower organisms like virus and bacteria which may multiply millions times or more in a day, and can afford to go in thousand wrong directions -- but not for higher organisms which produce a few off-springs in a year.

(We might be going in circles now...)

[Arete]

50 minutes ago, Jay Kulsh said:

It seems to me, this "process of selection" rationale works only for lower organisms like virus and bacteria which may multiply millions times or more in a day, and can afford to go in thousand wrong directions -- but not for higher organisms which produce a few off-springs in a year.

There are countless observations of natural selection in so called "higher organisms". Many feature as model organisms for natural selection - stickleback fish, toads, anoles, killifish, guppies, monkeyflowers, grasshoppers, Drosophila, jellyfish, lycophytes, to name a few. There are dozens of well known manipulative experiments that have comprehensively demonstrated natural selection in populations of relatively long lived, multicellular organisms, and thousands of population genetic studies on natural populations which do the same.

Natural selection in modern humans is also directly observed. The best example that immediately comes to mind is the Framingham Heart Study.

I think the concept that you're missing is that selection is dependent on population size. In small populations, genetic drift can overwhelm selection, leading to the loss of beneficial alleles and the fixation of deleterious ones. Conversely, the larger the population size, the lower the selection coefficient required to lead to fixation or extinction of a given mutation. https://www.nature.com/scitable/knowledge/library/natural-selection-genetic-drift-and-gene-flow-15186648/

[Jay Kulsh]

Arete, you sure provide good links. It will take me a while to go through them.

Thanks.

[Mark78L]

In support of Jay's enquiry:

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

This paper highlights the mutational bias generated by epigenetic stimuli associated with the plant Arabidopsis thaliana, which indicates that mutational evolution is not random, but a conscious response to improve survival on the part of the organism.

To use a crude example - indigenous people who live at altitude have developed a mutation that increases their ability to operate more effectively in low oxygen environments. Was this a random mutation that then led these people to colonize areas of high altitude (the mutation drove their behaviour), or, while living at altitude for extended periods did their genes adapt to the environment to improve their chances of survival? (the mutation was a direct response to environmental cues aimed at improving survival). The former equates to a random mutation, the latter a conscious adaption by the organism (mutation) in response to local conditions and the need to out-compete other organisms for survival.

The latter makes more sense to me.

We can't discuss evolution without considering consciousness. I would argue that mutations in genes that confer an advantage are primarily a conscious response to stimuli at the cellular level, or at a level of crosstalk between cells, mitochondria and the microbiome of larger organisms that all work symbiotically to, again, enhance their survival.
 

[Genady]

27 minutes ago, Mark78L said:

... while living at altitude for extended periods did their genes adapt to the environment to improve their chances of survival? (the mutation was a direct response to environmental cues aimed at improving survival). ... the latter a conscious adaption by the organism (mutation) in response to local conditions and the need to out-compete other organisms for survival.

"Genes adapt"? Why not a random change?

"Mutation a direct response"? Why not a random change?

"Conscious adaptation"? Why not a random change?