The "randomness" of evolution

[Ygggdrasil]

A few notes:

But that doesn't address all the issues and I am confident that other people have augmented that basic theory with 'add-ons' and really I don't know what they are. Can biologists here tell us what those are?

As noted, while evolutionary theory certainly originated with Darwin, you are correct that his thoughts on evolution are quite old (they pre-date knowledge of genes and genetics, which are crucial to our modern understanding of evolution). If modern evolutionary theory has a name, it would be generally referred to as the "modern synthesis," referring to a theory that synthesizes Darwin's model of natural selection with our modern understanding of genetics and molecular biology. A brief description of the modern synthesis can be found here: http://www.talkorigins.org/faqs/modern-synthesis.html

1. In answer to the OP's point, I think one of the points that has to arrive is that, surely, the theory is actually 'survival of the generally fit' rather than the fittest. I mean, if mutation is random, as the OP proposal (which seems to make sense to me), this will give a range of characteristics, some of which may not be particularly beneficial at a given point in time. But as environments and predation changes those 'hidden' characteristics may manifest and all of a sudden within the generic mix of those individuals, those that 'had been' the fittest all of a sudden, possibly instantaneously, die out. For example, say a species living in an arid area has some of its individuals 'uselessly' born with webbed feet but it doesn't disadvantage them too much and it just happens to be a characteristic, some have more webbed feet than others. The fastest runners don't have webbed feet and so do a bit better and slowly the population are moving towards 'no webbed feet', but still it is not such a disadvantage that those individuals die out. .... and then their habitat floods ...! So I think the substance of this is not merely 'the fittest' and 'survival' but actually if a species can successfully maintain the widest set of random characteristics then it is more likely to survive. As such, this is not survival of the fittest individuals, this is survival of the species with the widest set of tolerances to environmental changes. I am sure this has been proposed by evolutionary biologists and already has a name, which would address the OPs question.

It is important to remember that the fitness of an organism is a function of its environment. An organism that is the most fit in one environment is not necessarily the most fit in a different environment. Sometime, specialization to gain fitness advantages in one particular environment may cause fitness disadvantages in a different environment (e.g. the example of the cost of antibiotic resistance genes in an antibiotic free environment which was mentioned earlier). These trade offs between fitness and "robustness" or "evolvability" (i.e. ability to withstand environmental change) have been experimentally observed in laboratory models of evolution, for example in the following studies:

Stiffler, Hekstra and Ranganathan (2015) Evolvability as a Function of Purifying Selection in TEM-1 β-Lactamase. Cell 160:882 2015 https://www.sciencedirect.com/science/article/pii/S0092867415000781?via=ihub

Johnson et al. (2019) Higher fitness yeast genotypes are less robust to deleterious mutations. bioRxiv. https://www.biorxiv.org/content/10.1101/675314v1

3. Rate of change of species; A common ancestor of ours to other primates (as far as I know) is the Miocene Proconsul (https://www2.palomar.edu/anthro/earlyprimates/early_2.htm) which is said to have lived from 21 to 14 million years ago. But if it took 7 million years to become a new species then why is that an unrepresentative timescale for the 14 million years since then? I mean, surely there are more than two evolutionary 'species' steps from proconsul ape to us? Maybe not? I don't know? What do updated evolutionary theories say about this? I mean, taking that thought further, if we look at the first 'modern humans' 250,000 years ago, if one was here today they could integrate into society, we'd be able to have offspring with them and (presumably with training!) talk together. Yet that is 'only' 1/40th or so between proconsul ape and us, and it looks to me like there are going to be more than 40 times the number of differences between early homo sapiens and proconsul ape? Again are there any modern theories to explain what would initially appear to be, presumably, sudden accelerations in changes of species?

The rate of evolution is definitely not linear and there are periods (especially those associated with widespread environmental changes) that are associated with much higher rates of speciation (e.g. the Cambrian explosion). This relates to the previous point about the role of environment in determining fitness. Changes to the environment will alter the relative fitness of the populations in an ecosystem, precipitating changes in the populations in that environment. The idea that evolutionary change can happen in intense bursts of change followed by relative periods of stability is known as punctuated equilibrium (https://en.wikipedia.org/wiki/Punctuated_equilibrium). The extent to which "evolution by jerks" (punctuated equilibrium) vs "evolution by creeps" (gradualism) describes most evolutionary change is a subject of debate in the field.

 

[FactChecker]

So basically, life not only has evolved adaptive traits for species in their environments, but it has also evolved the process of evolution itself in a way that makes constructive mutations more likely, by making sure that important areas of DNA are more likely to be mutated.

How would the DNA know which parts to protect and which should it should allow to be changed for the better? IMHO, duplication would just slow down the process of evolution -- both for good changes and for bad changes. That might be a great weakness compared to organisms that evolve more rapidly for the better. I think that evolution would eliminate this weakness.

 

[Mary Conrads Sanburn]

Excerpt from the National Science Foundation:

Human Evolution's Winding Path

“Tim White is a world renowned paleoanthropologist and professor of Integrative Biology at the University of California at Berkeley. His work frequently takes him to study sites in Afar, Ethiopia, Jordan, Kenya, Malawi, Tanzania (at Olduvai Gorge and Laetoli) and Turkey. His primary research involves human evolution in all its dimensions and he and his colleagues are credited with the discovery in Ethiopia in 1995 of perhaps the oldest known human ancestor, Ardipithecus ramidus, dated to 4.4 million years ago. The National Science Foundation supports his work on a Middle Awash research project in Ethiopia.”

https://www.nsf.gov/news/special_reports/darwin/textonly/anthro_essay2.jsp

 

[Ophiolite]

Excerpt from the National Science Foundation:

Human Evolution's Winding Path

“Tim White is a world renowned paleoanthropologist and professor of Integrative Biology at the University of California at Berkeley. His work frequently takes him to study sites in Afar, Ethiopia, Jordan, Kenya, Malawi, Tanzania (at Olduvai Gorge and Laetoli) and Turkey. His primary research involves human evolution in all its dimensions and he and his colleagues are credited with the discovery in Ethiopia in 1995 of perhaps the oldest known human ancestor, Ardipithecus ramidus, dated to 4.4 million years ago. The National Science Foundation supports his work on a Middle Awash research project in Ethiopia.”

https://www.nsf.gov/news/special_reports/darwin/textonly/anthro_essay2.jsp

Would you elaborate on how this article is relevant to the specific questions the OP has? It's not clear to me.

 

[Mary Conrads Sanburn]

Excerpt from article Science Direct

Highlights


Randomness effects in the evolution of tag-based cooperation were studied.
•
An agent-based evolutionary model with memory, heritable traits, and asexual reproduction was employed.
•
Competition between dominant cooperating strategies (ethnocentrism and altruism) was analyzed.
•
Stochastically established indiscriminate altruism can permanently outweigh ethnocentrism.
•
Randomness plays an important role in promoting non-ethnocentric cooperation.

https://www.sciencedirect.com/science/article/abs/pii/S0376635715000042?via=ihub

 

[Ophiolite]

Excerpt from article Science Direct

Highlights


Randomness effects in the evolution of tag-based cooperation were studied.
•
An agent-based evolutionary model with memory, heritable traits, and asexual reproduction was employed.
•
Competition between dominant cooperating strategies (ethnocentrism and altruism) was analyzed.
•
Stochastically established indiscriminate altruism can permanently outweigh ethnocentrism.
•
Randomness plays an important role in promoting non-ethnocentric cooperation.

https://www.sciencedirect.com/science/article/abs/pii/S0376635715000042?via=ihub

Well, those are certainly relevant, however none of them appear in your original linked item which is not from Science Direct, but from the National Science Foundation.

 

[fresh_42]

Please stay on topic and consider the references. Just because a publisher has scientific textbooks in their portfolio doesn't mean that any publication is peer reviewed. A discussion about the seriousity of a publisher is not necessary and misleading. Individual journals count, not the company.

 

[BillTre]

IMHO, duplication would just slow down the process of evolution -- both for good changes and for bad changes. That might be a great weakness compared to organisms that evolve more rapidly for the better. I think that evolution would eliminate this weakness.

Gene duplications, and on a larger scale, genome duplications in particular have provided important material (usefully structured sequences) for evolution to modify and use. As a result, they let evolution happen faster in many cases.

Whole genome duplications have occurred several times, in several lineages during evolution.
In the evolutionary lineage leading to use (humans (and related animals)), two obvious whole genome duplications have occurred, after the branch point between the amphioxus (kind of like a fish without a head (its head is actually there, just not very distinctive from neighboring axial body regions)) and more familiar vertebrate animals (like fish, amphibians, reptiles, mammals, primates, humans).

Complex sets of genes (Hox gene clusters), that have been identified (by sequence) and seem appear to be obviously the result of whole genome duplications.
There clusters of genes contain several genes with several properties that vary colinearly with their position in the overall cluster.

• body areas (along the anterior-posterior body axis) are effected by mutations at colinear positions within the cluster
• genes are expressed in different anterior-posterior locations in the body
• Transcription usually (but not always) goes in the same direction along the DNA

The gene clusters, clearly show a history, in the vertebrate evolutionary lineage, of two genome doublings, followed by the different genes and clusters of genes taking on different regulator functions (often more anatomically refinded within the embryo)

About 20 years ago, NeoFunctionalization and SubFunctionalization was proposed to explain why genome duplications and single gene duplications could be useful in evolution, briefly:
All genes duplicated:

• lots of extra new copies of all genes
• Continued selection for the traditional gene function (of the original parental gene, in the unduplicated genome) should maintain one copy of the gene.
• The other copy is available for evolutionary tinkering (without strong selection to preserve its original function or prevent its elimination). This copy would be under no selective pressure to not change and variation could arise.
• If the modified version of the gene develops some function that is distinct and useful, it could be selected for and be maintained (over the generations). In development (where Hox genes work), this can involve things like: some sequence change, expression in different (perhaps subsets of) anatomical locations or at different times.

 

[mswzebo]

cmb - interesting thoughts. Some of this is actually far more fascinating than you are tempted to imagine.

First, the sequences like CAGCAGCAGCAG (simple, repetitive sequences) are in fact more likely to mutate. This is because of two processes called replication slippage and strand mismatch crossover which occur during DNA replication. However, these sequences are almost always found in regions that are transcriptionally silent - the DNA is never made into RNA, and there aren't any observed consequences of the mutations - so they don't contribute to changes in fitness. Perhaps these higher mutation rates are tolerated evolutionarily because there aren't any consequences so there is no pressure to fix it (at least in plants and animals, with huge genomes most of which are transcriptionally silent; bacteria don't have these repetitive sequences because their genomes need to be replicated rapidly so there is selective pressure to get rid of nonfunctional regions).

TOTAL ASIDE NUMBER ONE: sometimes they do matter - in fact- there is a whole set of diseases that are called trinucleotide repeat disorders. These include Huntington's and Fragile X syndrome. Here, the changes that occur are increases in the number of repeats...which increases the likelihood that offspring will have even more repeats...and the more repeats you have the more likely your cells are to methylate and transcriptionally silence the gene...which is ultimately the cause of the diseases.

BACK TO OP/OR MAYBE TOTAL ASIDE NUMBER TWO: Mutations are generally thought to happen at random - ie the probability of a mutation happening is independent of the effect on fitness. However, in 1988 John Cairns published a paper saying sometimes this wasn't true, at least in E. coli. There is an interesting, but rare, set of subcases where certain mutations will happen at higher rates in genes that have stopped working normally. Here is a review paper of this very cool phenomenon:

https://www.ncbi.nlm.nih.gov/pmc/articles/PMC2747772/
From the review:

"In a stimulating but controversial paper published in 1988, Cairns et al. (1988) described experiments suggesting that a population under selection had access to a process that could direct mutagenic change to the very genes that would relieve the selective pressure. Such a process, of course, would be a great boon to adaptive evolution (Fitch, 1982). After nearly 20 years of research, evidence now suggests that various types of stresses induce responses that have mutagenic consequences, and that sometimes this essentially random process can appear to be directed. Originally it was also thought that there was just one mechanism for all cases of adaptive mutation. This idea also proved to be wrong–it seems there are many mechanisms by which genetic change can be produced when organisms are under stress.

This review concentrates on stress-induced mutagenesis in Escherichia coli and related bacteria with a few examples from other organisms."

But, the phenomenon is simply not widespread enough to claim that it is a major force in evolution, which does in fact seem to be random and undirected.

OK, but TOTAL ASIDE NUMBER TWO OR THREE: Sometimes the creation of a new species happens instantaneously. This happens ALL THE TIME in plants, and is thought to be the case for something like half of extant flowering plants. The mechanisms have to do with polyploidy and hybridization. The classic example was Hugo de Vries, who found an evening primrose in his greenhouse that was bigger and more vigorous than the other evening primrose he had...and ultimately he figured out that it had twice as many chromosomes as the other plants and couldn't interbreed with them, but was self-fertile. Voila! Anew species in 1 generation. Check out autopolyplody and allopolyploidy as search terms for other examples.

OK BUT THE FLIPPIN COOL THING IS THAT THERE ARE HUMAN EXAMPLES TOO! Via an unrelated chromosome change mechanism (Robertsonian translocation) there can be instantaneous human speciation and there is actually an example of it having happened within the last ten years. Link to paper is here:
http://www.alliedacademies.org/articles/case-report-potential-speciation-in-humans-involving-robertsoniantranslocations.html
Ok, sorry about all the allcaps but cool stuff.

 

[Drakkith]

OK BUT THE FLIPPIN COOL THING IS THAT THERE ARE HUMAN EXAMPLES TOO! Via an unrelated chromosome change mechanism (Robertsonian translocation) there can be instantaneous human speciation and there is actually an example of it having happened within the last ten years. Link to paper is here:

From the paper:

Although a Robertsonian translocation carrier has a full genetic complement, their fitness is reduced due to high probability of genetically imbalanced gametes. However, this type of translocation can provide material for human evolution. Long term isolation of a group of individuals who are homozygous for a particular Robertsonian translocation chromosome could theoretically lead to the establishment of a new human subspecies having a full genetic complement in 44 chromosomes.

So there's not really an instantaneous speciation event in humans. It would take the isolation of a population with this type of translocation to create a new species/sub-species.

 

[jim mcnamara]

Long term isolation of a group of individuals

This is the definition of an isolated population, the basic "unit" of the process of speciation. Isolation can be one many things, some examples:
physical - ex: season or time of day for flowering, incompatible pollen/stigma, change in estrus pheromones, coloration.
geographic - ex: on an island, surrounded by glaciers in a valley, in a dry zone surrounded by a swamp.
behavioral - male displays, mating calls,
genetic - polyploidy (different numbers of chromosomes), parthenogenesis

For all of these speciation is also a function of generation time - hours like a mayfly, 18 years like humans.

And the most important point: our definition of species is largely arbitrary. Lots of exceptions and edge cases. It is an idea with great convenience for us humans cubbyholes to classify things. Nature doesn't give a darn about convenience.

Normally we think of species as unable to interbreed and produce fertile offspring - like mules = horse interbred with ass

Example exception: modern European humans have a tiny part of the Neanderthal genome, Asian humans a tiny part of the Denisovan genome. So, if those earlier humans were truly separate species, then why did mating with modern humans create viable offspring? Were they truly separate species or not, is the crux of the matter. Part of the answer is that evolution of the genus Homo is "bushy", not linear at all.

Really good article:
https://www.learner.org/courses/biology/units/humev/experts/tattersall.html

 

[Wellwisher]

Random mutations defy common sense. As an analogy and possible home experiment, open the code to the operating system of your computer. We will use that simulate the DNA codes for the cell. Make some random changes. You do not even have to know any coding to cut, paste, delete and move things, randomly. We will have an expert in coding play the role of natural selection, keeping track of the positive improvements you randomly create.

What will happen is the operating system will get bugs and then stop working way before you do anything useful with a random cut and paste approach. Life should have ended long ago if it allowed internal random mutations. This approach would mess things up faster than it could rebuild and/or improve. The fact that life persists and evolves implies life is an expert in code, making strategic changes in the code, that are planned in advance.

Random is an illusion that appear to work when we do not understand how things work in a logical fashion. This assumption is a philosophical artifact of using statistics. In statistical methods, we place phenomena in a black box; in the dark, and use the assumption of random because we are blind folded by the black box. We monitor input and output and analyze these based on the assumptions of a mathematical method. What has happened is the black box approach, which is still very useful, has been reversed engineered into a philosophical assumption of reality; chicken or the egg.

Here is the logic behind mutations. If you look how life formed from simple chemicals; abiogenesis, life, from the very beginning evolved in water. Water was the basis for natural selection, at the nanoscale, since day one. This can be inferred from the observation that no protein, RNA or DNA or combination, will work properly in any other solvent besides water. Everything is tuned to the unique potentials and properties created by the water environment, just as animals and plants are tuned to their specific macro-environments.

Mutations are a form of natural selection by water, at the nanoscale. Water has a long term goal; based on free energy, and is responsive in real time to external changes that alter the aqueous environment down to the nanoscale.

For example, the DNA is the most hydrated molecule in the cell. This ranking, as the organic molecule, with the closest free energy relationship with water, allows water and the DNA to mutually interact like brothers. That mutual benefit, for water, was one goal of abiogenesis; RNA and DNA were destine to happen because of the needs of the water environment.

 

[Drakkith]

@Wellwisher
Mutations are random because they are caused by unpredictable events. UV damage, oxidative damage, translocation, deletions and errors during replication, these are all random. However, the selection process that determines which mutations are beneficial, neutral, or detrimental is not random. That is deterministic.

Typing from my phone right now, so I can't elaborate at this time.

 

[BillTre]

What will happen is the operating system will get bugs and then stop working way before you do anything useful with a random cut and paste approach. Life should have ended long ago if it allowed internal random mutations. This approach would mess things up faster than it could rebuild and/or improve. The fact that life persists and evolves implies life is an expert in code, making strategic changes in the code, that are planned in advance.

Computer code can be use as a decent metaphor for certain aspects of life, but not in this way:

1. Some code can be scrambled and still work, depending how it is done.
2. Living things are much more robust than a brittle bit of machine code that will break for lack of the smallest part.

Random is an illusion that appear to work when we do not understand how things work in a logical fashion. This assumption is a philosophical artifact of using statistics. In statistical methods, we place phenomena in a black box; in the dark, and use the assumption of random because we are blind folded by the black box. We monitor input and output and analyze these based on the assumptions of a mathematical method. What has happened is the black box approach, which is still very useful, has been reversed engineered into a philosophical assumption of reality; chicken or the egg.

Mutations are actually pretty well understood. There are many kinds. They have different causes. You explanation does not seem to accept them.

Mutations are a form of natural selection by water, at the nanoscale. Water has a long term goal; based on free energy, and is responsive in real time to external changes that alter the aqueous environment down to the nanoscale.

You seem to be confusing mutations and natural selection (in which case, see here of here) or have your own definition of what they are and forget to tell us what that is (in which case it would be a personal theory which the Physics Forums does not permit).
If water is the basis for selection, why is not all life the same? Water is largely the same everywhere, so the selection should be also and all life should have very similar characteristics.

You are making some rather extraordinary claims.
They should be referenced in some why if you want them to be taken seriously.

"extraordinary claims require extraordinary evidence": from Carl Sagan and others

 

[Drakkith]

Random is an illusion that appear to work when we do not understand how things work in a logical fashion.

That's certainly one way of looking at randomness in the framework of a completely deterministic system. In such a system a 'random' event would be one which, although in principle capable of being predicted, is simply the result of too many events and variables to predict in practice.

However, the universe is not deterministic. Quantum effects, such as the decay of atoms, the emission of radiation, and the reaction between one molecule and another are all partially or entirely random events since they cannot be entirely predicted in advance. Not even in principle.

Mutations are a form of natural selection by water, at the nanoscale. Water has a long term goal; based on free energy, and is responsive in real time to external changes that alter the aqueous environment down to the nanoscale.

Mutations are not natural selection. The two are fundamentally different things. You might be able to argue that water contributes to natural selection in the micro-environment, but saying that mutations are a form of natural selection is simply wrong. Mutations are changes in the DNA/RNA of an organism and natural selection is the differential survival and reproduction of individuals due to differences in phenotype.

 

[Wellwisher]

If water is the basis for selection, why is not all life the same? Water is largely the same everywhere, so the selection should be also and all life should have very similar characteristics.

Life does have characteristics that are the same, such as DNA, protein, membranes, etc. This is commonality at the nanoscale. These are optimized to the water environment. The difference at the macro scale, as you point out, is impacted by the character of the accumulating organics in mother cells. These create unique potential in the water. The water will lower potential via cell cycles, whereby change can occur.

An easy way to see this is with the system of water and oil. If we shake this system vigorously it will form an emulsion. The water gains potential via surface tension. The water lowers the potential by phase separating the randomness of the emulsion into order. Water can maximize it self by self bonding and segregating organics.

The mother cell is accumulating reduced materials, which increases the potential of the water. She adds the oil and emulsion affect by expending energy. The cell cycle is a way for the water to remove the potential lowering free energy through changes in entropy and enthalpy. The entropy increase is where many changes will occurs. The mother cell sets the water up to assist her in change. Directed mutation is connected to entropy being a state variable. A very specific state defines a very specific amount of entropy. This is not random. For example the entropy of water at 25C and one atmosphere is a constant measured the same by all labs. Some Mutations are state variables.

 

[Drakkith]

I feel that your description is extremely limited and misses a great deal of what makes evolution work. The properties of water and how it interacts with other molecules is important, but it is not the fundamental aspect of natural selection or evolution. The general principles of both of these are independent of the exact details of how life functions. All it requires is that there are heritable changes in traits and that these traits affect how well an organism can reproduce and pass on these traits.

 

[jim mcnamara]

@Wellwisher - Please cite some real journal or textbook articles that support your statements. For example, IMO, you do not seem to understand random in terms of mutations. Your comments look like speculation. PF is not the place for that, so please cite articles or at least commonly known facts from classwork. Thank you.

 

[Wellwisher]

The problem with the notion of random mutations, is that a truly random mutation process will do more harm than good, over time. I originally presented an example of randomly cutting, pasting, deleting and inserting random computer code into the operating system of your computer. Your computer will crash long before you randomly come up with something that is useful. Life does not crash this way, but rather life progresses and even improves. Random is not exactly correct.

I think I figured out why this is so hard to see using contemporary wisdom. Consider a deck of cards. I will randomize the deck of cards by shuffling it. If I then played solitaire, although the cards in the deck has been randomized, every card in that deck leads to a positive outcome. This positive outcome may not happen right away, but every card has a place, in terms of game play. Although the cards are randomized before I start, the shuffle plus the rules of play does not end in a purely random event. The rules of the game have loaded the dice, so to speak, with 52 positive outcomes, but no purely inert and detrimental outcomes. Random should contain all these options, but the random deck was stacked by the rules of the game.

Say I added a dozen or so inert cards to the deck, that do not lead to positive outcomes but will more to less delay the speed of the game. I will also add a dozen negative outcome cards which can bottleneck the game or reverse what I have done. Now the odds for me ever finishing the game, go way down. Both the original deck and this new modified deck are shuffled and randomized, but the outcome is different for each deck. The second deck leads to a more random result regardless of the rules.

What appears to be random, in terms of mutations, is analogous to loaded dice. These dice are not overlay loaded, thereby allowing all sides to appear. However, they will not appear at same rate. Water is how the cell loads the dice, so the randomness in the shuffle has a net positive outcome over time.

If you look at dice, these are artificial and not found in nature. We, as humans, purposely weigh all sides to be the same, so we can create a random outcome. This is not how the forces of nature work, for example. Each side of the natural force dice carries a different weight. The differences in the manmade dice is based on a superficial facade, that does not affect the weight of the sides. Nature loads the side of the dice, based on the lack of symmetry in the various forces in the facade called atoms. Water has the best dice load.

In the late 1950's it was discovered that protein fold with exact folds. Up to that point, it was assumed that proteins folded based on average folds, due to thermal vibrations and other random events. Over 60 year later, there is still no good statistical explanation. The explanation has to do with water loading the dice, to where the odds become 1.0.

Someone may need to develop a new type of statistics called loaded dice statistics; LDS. The load is based on the laws of physics as applied to various atoms and molecules.

 

[FactChecker]

The problem with the notion of random mutations, is that a truly random mutation process will do more harm than good, over time. I originally presented an example of randomly cutting, pasting, deleting and inserting random computer code into the operating system of your computer. Your computer will crash long before you randomly come up with something that is useful. Life does not crash this way, but rather life progresses and even improves. Random is not exactly correct.

I think this is the wrong conclusion. Suppose there are many random mutations that cause premature deaths in a population for every mutation that gives an improvement. That still leads to an improved evolution. The surviving ones are the ones that count.

 

[Bandersnatch]

The problem with the notion of random mutations, is that a truly random mutation process will do more harm than good, over time. I originally presented an example of randomly cutting, pasting, deleting and inserting random computer code into the operating system of your computer. Your computer will crash long before you randomly come up with something that is useful. Life does not crash this way, but rather life progresses and even improves. Random is not exactly correct.

You're confusing individual organisms with populations. Mutations happen to individuals. Evolution happens to populations.
Individual life crashes all the time - for example, an embryo that is not viable due to a sufficiently harmful mutation in its genetic code will spontaneously abort. Such mutations are eliminated from the population at the onset.
According to this source: https://books.google.pl/books?id=4Sg5sXyiBvkC&pg=PA438&redir_esc=y#v=onepage&q&f=false
30 to 40% of all conceptions miscarry (even if not all should be attributed to mutations in the genetic code).
Furthermore, the mutations that are not harmful enough to outright kill the organism but merely reduce fitness (i.e. reproductive success) are eliminated from populations over time. There is no exactness here, just a selection process that eliminates detrimental mutations so that you only notice the 'good' ones.

 

[BillTre]

@Wellwisher, you have yet to provide any references in support of your personal theory about:

• how evolution works
• how mutations happen

You are creating extended mixed metaphors (cards and dice) which are unconnected with biology except through a confused claim that water in some way causes non-random mutations.
You seem to be starting with little or no knowledge of biology and do not appear interested in filling those gaps.
You are not really responding to queries from other posters other than telling everyone else they don't understand things.

Is there anyone else who shares these ideas with you (that means provide citations or links to publications of some kind) or are you just making them up on the fly?

 

[Drakkith]

@Wellwisher your example using computer code is flawed. A better analogy would be an assembly line of computers all getting random changes in their code as they are built. The ones with serious errors crash immediately and are discarded. The ones that work go on to be sold. Some of the latter have code that is actually an improvement over the prior code. This better code can then serve as the template to program future computers.

In such an example computer systems would slowly improve over time, just like what happens in evolution.

 

[berkeman]

or are you just making them up on the fly?

That's kind of a mixed metaphor...

https://i.ytimg.com/vi/huDDaj0PjLU/maxresdefault.jpg

 

[BillTre]

OK @berkeman, now I get it.