and taking into account that increasing complexity in organisms generally means complexity in phenotype, it should be the case that more complex the organism, the greater its genome size. — TheMadFool

I don't have time to go into detail on Genetic Science, so I'll just mention that in Systems Theory, Reductive complexity (sheer numbers) is distinguished from Holistic Complexity (interrelationships). Reductively & numerically, a pile of sand may be "complex" (thousands of grains), but add a cement binder (links between grains), and the resulting concrete is a holistic system that is much stronger, and more complex, than its component parts.

Early Geneticists, based on their assumption that simple numerical size of the genome was the most important factor in the resulting physical & behavioral expression, discovered that some dumb simple organisms had more genes than homo sapiens. So the intelligence aspect of the human phenotype must derive from something "more than" the sum of its parts : Holism. The "more than" is the links (interrelationships) between isolated parts, e.g organization. :smile:

Holism : https://en.wikipedia.org/wiki/Holism_in_science

DNA and RNA mechanics underlay all species differentiation. The need to develop an idea like ecology is necessary to the discovery of that process. We understand things happening in a certain situation. We don't know what that situation is.

:up:

:chin:

Does Genotype Truly Determine Phenotype?
Genotype = The genetic composition of organisms

Phenotype = The set of observable characteristics of an individual [[b]resulting from the interaction of its genotype with the environment.[/b]] — TheMadFool

Seems to me that you answered your own question in your OP. The phenotype is determined by not just the genotype, but by the environment as well. Think of phenotype as a feedback loop between genotype and the environment.

The phenotype of prey is determined by the phenotype of other predators and the predators are determined by the phenotype of prey, just as the phenotype of some insects and plants are intertwined.

I don't have time to go into detail on Genetic Science, so I'll just mention that in Systems Theory, Reductive complexity (sheer numbers) is distinguished from Holistic Complexity (interrelationships). Reductively & numerically, a pile of sand may be "complex" (thousands of grains), but add a cement binder (links between grains), and the resulting concrete is a holistic system that is much stronger, and more complex, than its component parts. — Gnomon

Right, the grains and the medium binding the grains is a more complex system than just grains. We can't talk about things without talking about how those things interact with other things or are determined by other things, like the environment it is part of. It's not that it is more complex, it is that it impossible that objects can be apart and not determined by the very environment it is part of, and all objects are parts of environments.

Considering the scientific consensus that genotype determines phenotype and taking into account that increasing complexity in organisms generally means complexity in phenotype, it should be the case that more complex the organism, the greater its genome size. — TheMadFool

A couple of things to be aware of:

1. Genetics doesn't directly define the qualities of an organism; there is an interplay with the environment and the rest of the organism. We're aware of nature vs nurture in terms of personality, but in fact all of our physical characteristics too depend on a complex interplay within the body, within the womb environment and then later within the external environment.
To give a trivial example, no genome is big enough to code for the structure of the capillaries. Instead, a process happens that results in a certain "branchiness" of capillaries. But "run" the same genome again, and the capillaries will not be the same.

2. The more important point; there's an awful lot of redundant genes; the majority of the human genome is "junk" DNA and doesn't code for any protein i.e. it does nothing. In certain plants that have a much bigger genome than ours, the proportion of junk or duplicate DNA is much greater (duplicate DNA can sometimes have a different phenotypical effect than one copy, but often is redundant).
So there is no inconsistency here.

I find it hard to believe that the environment determines phenotype in an immediate sense i.e. in real time. Perhaps as part of evolution - a slow process requiring multiple generations - the environment can influence phenotype through natural selection. Note that in this context (evolution/natural selection) the phenotypic complexity is coded into the genotype i.e. the information for complexity is contained in the genes. The environment can only produce complexity in phenotype through its effect on the genotype.

You seem to be saying the same thing as Harry Hindu above.

After what I've said above in my reply to Harry Hindu, we must return to the original statement regarding phenotype - that it's determined by genotype.

I've answered the OP with demonstrable scientific facts, so I'm not sure what there is left to debate and why you still find it hard to believe. To reiterate once again simply:

1. There is not a one-to-one correspondence between genes and phenotypal complexity: e.g. the layout of blood vessels in the body is extremely complex, and it isn't coded directly in the genes.
2. Most DNA does not code for anything, or redundantly codes for the same protein. This is why there is no simple correlation between size of genome and complexity of the organism. Often the primary factor in how big a genome is is how many chromosome duplication events have occurred in an organism's history.

I've answered the OP with demonstrable scientific facts, so I'm not sure what there is left to debate and why you still find it hard to believe. To reiterate once again simply:

1. There is not a one-to-one correspondence between genes and phenotypal complexity: e.g. the layout of blood vessels in the body is extremely complex, and it isn't coded directly in the genes.
2. Most DNA does not code for anything, or redundantly codes for the same protein. This is why there is no simple correlation between size of genome and complexity of the organism. Often the primary factor in how big a genome is is how many chromosome duplication events have occurred in an organism's history. — Mijin

Thank you. My two cents:

1. I'm intrigued by the fact that no two vascular trees are identical. However is a vascular tree an instance of complexity? Granted that, as I said, no two are identical but a bird's eye view of the vascular system bespeaks a simplicity - a result of a simple random branching algorithm. Nothing that's beyond the capacity of genomes.

2. As you said, there's some parts of our genome that don't code for any protein and are labeled as junk DNA but this is, as is the case in all of science, only the current best judgement on the matter. It's liable to change as new evidence comes to light. If it turns out that junk DNA truly has zero phenotype information content then it speaks in my favor - genotype is not sufficient to explain phenotypic complexity.

You should probably consult the field of epigenetics.
Genes may or may be expressed depending on environmental and other influences.
So genes are not destiny. There is an interplay between your genetic code and the environmental experience which determines phenotypes and gene expression and interplay.
Your genetic clone would not be you, having had a lifetime of different experiences and influences.

You should probably consult the field of epigenetics.
Genes may or may be expressed depending on environmental and other influences.
So genes are not destiny. There is an interplay between your genetic code and the environmental experience which determines phenotypes and gene expression and interplay.
Your genetic clone would not be you, having had a lifetime of different experiences and influences. — prothero

Right! There's epigenetics but aren't you forgetting the fact that genes that function epigenetically are just the tip of the iceberg? What about the genes that control basic life-sustaining biochemical/physiological function? Differences in complexity can be seen in these too. The essential life-support system of a human is far more complex than that of a bacterium and the genes that determine it are necessarily expressed.

Given what I said in the preceding paragraph and it's true that genotype determines phenotype, it must be that more complex organism should possess larger genomes and more genes. That, according to Nature article I linked to in my OP, is not true.

Given what I said in the preceding paragraph and it's true that genotype determines phenotype, it must be that more complex organism should possess larger genomes and more genes. — TheMadFool

Rather than genes determining everything, it is better to say that they constrain it.

To determine everything suggests a one to one relationship - one bit of genetic information to fix one bit of phenotypic physical structure.

But as others have said, far less information is needed to simply constrain physical growth processes. The genes can switch growth phases on and off. The growth itself can then be random - as in the fractal branching patterns of a capillary network.

Or rather, the information can be considered part of the environment - the constraints the physical context exerts.

So a capillary will make its decision to branch by sensing the degree of shear force being exerted by the blood flow. Genetics will set some general optimal value to aim at. The physics of what is happening on the ground creates the feedback signal which allows this value to be targeted by the branching.

The actual pattern of branching is not determined, just generally constrained by a global optimal value.

This is why the human brain could evolve so quickly from the ape brain. It was a simple matter of letting cell growth run on a few divisions longer. Just a small adjustment to the growth clock.

But then also, human brains were left more open to the influence of environmental information. As newborns, humans are especially helpless as the brain is still just laying down tissue. Even as teenagers, the higher areas of the cortex are still immature by the standards of other large brained animals.

So as a genetic trick, our nervous system is exposed to growth-based learning or adaptation for far longer before it gets myelinated - insulated and hardwired into place. The environment contributes much more to the final shape of the circuits. Which is important as humans want their brains shaped by another whole level of information - human linguistic culture.

Not that I am saying this is the reason for the basic paradox of the article. Folk are still searching for a solid reason why DNA doesn't more directly correlate with phenotype complexity.

I googled out of interest and the only hint seems to be that larger cell types in slower growing animals seems to correlate with chunky genomes. But not enough is known about what is actually junk DNA, what is epigenetic DNA, to answer why smaller or larger genomes might be evolutionarily favoured as different responses to different circumstances.

There is an open question to be answered here.

However the human brain is the most complex chunk of matter in the known universe in terms of its density of connectivity, especially when scaled for body mass. And also it is a prime example of the way that genes only need to regulate growth in a broad brush way. The bulk of the information can be provided by a tissue or organ learning to adapt its fit to its environment.

So phenotype expression is the product of nature and nurture always. Two sets of constraint acting to produce the outcome. Genetic information and environmental information - the environment being both the internal economy being experienced by the capillaries and the external world, and even culture, as it impinges on the body and nervous system.

:up: :ok:



What do you think of the clip? Is Neil deGrasse Tyson talking out of his hat?

What do you think of the clip? — TheMadFool

God awful frankly. :grin:

The fact that there is only 1% difference doesn't mean that a 2% difference would be just as dramatic.

The difference between apes and humans was the evolution of some small changes.

A key one was a reshaping of the vocal tract so that it could be used to make clearly articulated sound sequences - a proto-code in the sense that an alternating pattern of consonants and vowels is a super-effective medium for sending an informationally-structured message.

So apes have throats, tongues and lips. But the human growth schedule tweaked our vocal tract so that it could "bite" sounds off. Our vocalisation was digitised in that we could easily make 30 or 40 very brief, but easily distinguished, noises. And from there, language could take root.

It might have been a 1% level change to what the genes had to code for. But it unlocked another whole possibility of informational semiosis - the linguistic one that encodes human culture.

So what would a 2% change unlock in an alien? What else did deGrasse have in mind?

As it happens, in humans, words were topped by numbers. And speaking by writing. So that first innovation is still unfolding to reveal its further delights.

The 1% genetic change isn't done. And by the same token, the memetic change - our linguistic culture - may in turn have barely got started as the source of stored information shaping us as creatures.

Meanwhile back to the brain and what the 1% was responsible for there.

An important growth schedule change was in the degree of neural "top-downness". The spindle cell and mirror neuron type stories we read about. The human brain grows both larger due to its prolonged growth spurts and it also produces more neurons that feed downwards from higher to lower levels.

This is all part of the constraints-base logic of biology. The way to get smarter is to be able to apply more context to what is being experienced. The more you can predict what your sensory inputs ought to be, the less you need to try to figure out what surprising thing just happened.

And this applies to both sensory perception and motor control. So humans are more equipped for making and using complex tools. The 1% grew more top-down connections to control our hands and also our vocal cords.

But de Grasse is pretty rubbish here. He doesn't explain anything.

By chance, early homo evolved a slightly different vocal tract. Probably for making a wider variety of expressive emotional calls. And to go with that, there was a shift to higher level control over those vocalisations.

Apes make emotional grunts and whoops through the cingulate cortex - tightly tied to the limbic areas that do emotion. Humans over-ride the cingulate (except for involuntary swearing and the like) by shifting control up to the prefrontal where the vocalisation can instead be generated using the rational expressive structure of syntactical habits.

So early homo looks to have undergone a slight genetic set of tweaks to be able to make an interesting variety of crisply structured noises. Not yet meaningful in the sense of an actual language, but pragmatic in that it helped co-ordinate the intensely social life of a tool-using hunter-gatherer.

And then it is a small step to actually inventing a working grammatical structure, a first habit of communicating, and thus thinking, symbolically (rather than iconically or indexically).

We don't know it actually happened this way. But it is the reasonable speculation given data points like brain endocasts, the sudden appearance of symbolic art, the controversy over whether Neanderthals had hyoid bones, etc.

And certainly, it was the invention of symbolic and syntactic speech that was the revolutionary deal. A brand new encoding machinery to follow on from the long-established one of DNA.

Since you brought up language, there appears to be a way in which a smaller genome can produce greater complexity than a bigger one.

Imagine 4-bit genome (ABCD) and a 5-bit genome (EFGHI) and suppose that the 4-bit genome produces 2-bit sized proteins in which order matters (permutation) and that for the 5-bit genome produces 2-bit sized proteins in which order doesn't matter (combination).

Then we would have for the 4-bit genome the following protein family: AB, BA, AC, CA, AD, DA, BC, CB, BD, DB, CD, DC = 12 proteins.

For the 5-bit genome, the protein family: EF, EG, EH, EI, FG, FH, FI, GH, GI, HI = 10 proteins.

As you can see, the 4-bit genome (ABCD) that depended on permutation has a larger family of proteins (12) than the 5-bit genome (EFGHI) that used combination for its protein family (10).

Does it work like that?

Information theory proves that a "two bit" code can already encode infinite variety. A binary on/off switch is already capable of unlimited semiosis.

So a four bit genome vs a five bit genome essentially makes no difference in terms of its raw ability to encode information.

A binary code is all that is actually needed - at least for a physics-less notion of Turing computation or Shannon message transmission.

DNA then makes use of a "three bit" codon structure as it actually has to manipulate physics while doing its informational thing. It takes three DNA bases to spell out the name of a particular amino acid.

With four bases and a triplet codon ... well there are plenty of accounts of the story...

The three-letter nature of codons means that the four nucleotides found in mRNA — A, U, G, and C — can produce a total of 64 different combinations. Of these 64 codons, 61 represent amino acids, and the remaining three represent stop signals, which trigger the end of protein synthesis.

Because there are only 20 different amino acids but 64 possible codons, most amino acids are indicated by more than one codon. This phenomenon is known as redundancy or degeneracy, and it is important to the genetic code because it minimizes the harmful effects that incorrectly placed nucleotides can have on protein synthesis.

https://www.nature.com/scitable/topicpage/the-information-in-dna-determines-cellular-function-6523228/

So the general message would be that DNA had sufficient resources to combine enough amino acids in enough ways to be able to build an unlimited number of proteins. The information encoded could be turned into molecular machinery that self-assembled and started to physically constrain chemically-energetic processes.

The complexity that matters here is not the computational one (where a binary code would already be enough to represent every combination possible - and combination isn't actually complexity at all).

Rather it is all about the semiotic issue of being able to generate a sufficient variety of molecular machinery to regulate a metabolic world in the name of a general organismic goal (living, breathing, surviving).

Complexity from an adaptive systems perspective is all about the fact a semiotic relationship is in play - that encoded information is constraining physical dynamics.

If there is no semiosis happening, then the system isn't actually complex. Only complicated.

:up:

Does it work like that? — apokrisis

I don't know. I'm trying to find some way for a smaller genome to pack more punch than a larger one.

combination isn't actually complexity — apokrisis

Yes, I've been trying to get my hands on a good definition of "complexity" with no success.

I'm trying to find some way for a smaller genome to pack more punch than a larger one. — TheMadFool

The way to dig into that is consider the proteome. How many different proteins does any organism actually use? Do humans beat lungfish on that score?

Assuming a one-to-one correspondence between genes and proteins would mean that there are at least 20,000 proteins corresponding to roughly 20,000 genes for humans. The proteome can be larger than the genome, especially in eukaryotes, as more than one protein can be produced from one gene due to alternative splicing (e.g. human proteome consists 92,179 proteins[citation needed] out of which 71,173 are splicing variants[citation needed]).

On the other hand, not all genes are translated to proteins, and many known genes encode only RNA which is the final functional product. Moreover, complete proteome size varies depending on the kingdom of life. For instance, eukaryotes, bacteria, archaea and viruses have on average 15,145, 3,200, 2,358 and 42 proteins respectively encoded in their genomes.

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

Yes, I've been trying to get my hands on a good definition of "complexity" with no success. — TheMadFool

One good way to think about it is the simplexity vs complicity distinction that Stewart/Cohen made in their readable book, The Collapse of Chaos.

Algorithmic complexity is another more mathematical way of framing the issue - the search for the most compact program that could generate some particular bit string.

The 1990s saw this issue hammered out in a variety of ways. The Santa Fe school was famous for its Complex Adaptive Systems approach.

So there is a large literature on this core issue. Again, I take the semiotic approach of theoretical biologists like Howard Pattee. A complex system is described as one that has an "epistemic cut" - an internal informational model of what it ought to physically be.

So life and mind are complex systems on that score. Physics and chemistry can only produce complication.

The proteome can be larger than the genome, especially in eukaryotes, as more than one protein can be produced from one gene due to alternative splicing

That means there's no clear-cut definition of genes and ergo, genomes. Biology, unlike physics, appears to be more fluid. Perhaps the issue will be resolved once we define "gene" and "genome" in a better way.

Algorithmic complexity is another more mathematical way of framing the issue - the search for the most compact program that could generate some particular bit string. — apokrisis

Yes, I was thinking along those lines, wondering whether multifunctional swiss knives qualify as an instance of complexity. One problem though: it's generally believed that evolution evinces a progress from simplicity to complexity but if you take the idea of algorithmic complexity and apply it to the universe then, since the universe began, according to a science book, by fixing the value of just six numbers (referring to known physical constants), doesn't that mean the graph of complexity is showing a downward trend? After all there are more bits of information in our genome than in there are in just six numbers?

That means there's no clear-cut definition of genes and ergo, genomes. Biology, unlike physics, appears to be more fluid. Perhaps the issue will be resolved once we define "gene" and "genome" in a better way. — TheMadFool

Biologists knew that already. Cracking the genome was the easy bit. The hard work starts with working out how the information regulates the physics.

Yes, I was thinking along those lines, wondering whether multifunctional swiss knives qualify as an instance of complexity. — TheMadFool

A knife that is equally bad for every job? :chin:

One problem though: it's generally believed that evolution evinces a progress from simplicity to complexity... — TheMadFool

Again, this is where I admire the crystal clarity of Howard Pattee. He identified the epistemic cut as the definition of life. So even the simplest RNA soup counts as already irreducibly complex. There is just nothing in the physics that explains what is going on anymore. You have to see how information has now entered the room.

but if you take the idea of algorithmic complexity and apply it to the universe then, since the universe began, according to a science book, by fixing the value of just six numbers (referring to known physical constants), doesn't that mean the graph of complexity is showing a downward trend? After all there are more bits of information in our genome than in there are in just six numbers? — TheMadFool

The Big Bang was an ultimately simple state - a vanilla bath of boiling hot radiation. And the Heat Death will also have an ultimate simplicity - a vanilla bath of radiation so cold and thin that its just a rustle of zero degree photons.

It's the bit in between where complexity arises via a series of symmetry breakings. The Higgs field switched on mass and suddenly there were all these sluggish particles cluttering up the vacuum. The vanilla radiation bath became a cosmic dust bowl.

From there, it just got worse. The dust particles - hydrogen and helium atoms - clumped and caught fire. Stars emerged as fusion furnaces making the light elements. A reprocessing by super-novae then produced all the heavy elements to.

Next came planets and even planetary biofilms - at least on Earth there is life.

So yes, simplicity led to complexity. But simplicity gets to win in the long run.

1. I'm intrigued by the fact that no two vascular trees are identical. However is a vascular tree an instance of complexity? Granted that, as I said, no two are identical but a bird's eye view of the vascular system bespeaks a simplicity - a result of a simple random branching algorithm. Nothing that's beyond the capacity of genomes. — TheMadFool

Of course it's not beyond the capacity of genes; I'm not the one arguing that anything is.

The point is simply this: the genome is a recipe for how to make a human. It doesn't, and couldn't, encode everything about the end product.

As you said, there's some parts of our genome that don't code for any protein and are labeled as junk DNA but this is, as is the case in all of science, only the current best judgement on the matter. — TheMadFool

"Best judgement" is somewhat misleading here, making it sound like some best guess.
Actually our understanding is based on a number of factors and testable observations. We see that long stretches of DNA can be duplicated, and we see duplication events, such as whole chromosome duplication occurring in plants. We know what sequence of alleles will code for a protein, and which cannot result in a protein being produced. We can see sequences that were incorporated from retroviruses, and can see in which species they were first incorporated (this is very good evidence for evolution btw). So we can be very confident at this point that junk DNA is a thing, and that most DNA is junk (as most fits within the above and similar categories).

As with all of science, we could be wrong. But also as with all of science, we can talk of how confident we are in a model based on its testable predictions and inferences. We can have similar confidence in junk DNA as we have for, say, plate tectonics.

If it turns out that junk DNA truly has zero phenotype information content then it speaks in my favor - genotype is not sufficient to explain phenotypic complexity. — TheMadFool

I don't follow your logic.
But also, let's not get sidetracked; this point was addressing your point about how simpler organisms can have a bigger genome than humans. Well, this is the answer.

the genome is a recipe for how to make a human. It doesn't, and couldn't, encode everything about the end product. — Mijin

Looks like a contradiction to me. The human = the end product.

"Best judgement" is somewhat misleading here, making it sound like some best guess. — Mijin

Not really, judgement involves evidence, guessing doesn't.

So we can be very confident at this point that junk DNA is a thing — Mijin

No well-respected scientist will evince that much confidence in his/her findings.

As with all of science, we could be wrong. — Mijin

I take back what I said.

I don't follow your logic. — Mijin

If junk DNA is for real then genome size wouldn't reflect phenotypic complexity - there would be less genes than an organism's complexity would suggest. Please have a dekko at my discussion with apokrisis.

this point was addressing your point about how simpler organisms can have a bigger genome than humans. Well, this is the answer. — Mijin

I've had a lengthy and productive discussion with apokrisis who has considerably more knowledgeable than me on the issue.

There is just nothing in the physics that explains what is going on anymore — apokrisis

Biochemistry? Physiology? Can't life be viewed as a computer with different programs being executed on it? The hardware is the body and the software is the various functions the body's capable of, all encoded in the genome. Viewed in such a light, physics and chemistry is assuredly inadequate in providing an explanation for life but there's still hope since we have the "code of life" - our DNA - to catch our fall as it were.

The dust particles - hydrogen and helium atoms - clumped and caught fire. Stars emerged as fusion furnaces making the light elements. A reprocessing by super-novae then produced all the heavy elements.... — apokrisis

...and then, you open a Hotel.

I found an interesting theory - https://www.sciencedirect.com/science/article/pii/S1674205215001604

A likely reason for all the junk DNA is that DNA would become parasitic on itself. Individual segments would start copying and pasting themselves into the genome if they can get away with it. That would be just selection at work. The genome would become host to its own parasitic segments and so become bloated.

Then to explain why this happens to some genomes and not others, this paper argues that the genome has a delicate epigenetic network whose balance would be disturbed by this junk DNA inserting itself randomly. Mostly such defects would get edited out by evolution if the fitness of the genome was thus compromised.

But if a bit of parasitic DNA spawned a shower of copies, and these got inserted in a way that kept the global epigenetic network in balance, then the genome would still be fit and sudden bloating by multiple segments would be invisible to the forces of selection. The genome would be stuck with this parasitic load.

So it is a neat suggestion. The paper has a lot of good general background too.

Another relevant article is - https://www.scientificamerican.com/article/is-junk-dna-what-makes-humans-unique/

This talks about how the difference between ape and human brains is more about the epigenetic timing of cell division and growth schedules than the expression of any particles coded proteins.

Machine learning isn't always that great.
Could it be the case that the arrangement is so complex the pattern to which the phenotype is being acquired is invisible? There is a "mysterian" argument for consciousness maybe there is one for genetics too. Something in the process is invisible to us due to the way our cognition works.

I found an interesting theory - https://www.sciencedirect.com/science/article/pii/S1674205215001604

A likely reason for all the junk DNA is that DNA would become parasitic on itself. Individual segments would start copying and pasting themselves into the genome if they can get away with it. That would be just selection at work. The genome would become host to its own parasitic segments and so become bloated.

Then to explain why this happens to some genomes and not others, this paper argues that the genome has a delicate epigenetic network whose balance would be disturbed by this junk DNA inserting itself randomly. Mostly such defects would get edited out by evolution if the fitness of the genome was thus compromised.

But if a bit of parasitic DNA spawned a shower of copies, and these got inserted in a way that kept the global epigenetic network in balance, then the genome would still be fit and sudden bloating by multiple segments would be invisible to the forces of selection. The genome would be stuck with this parasitic load.

So it is a neat suggestion. The paper has a lot of good general background too.

Another relevant article is - https://www.scientificamerican.com/article/is-junk-dna-what-makes-humans-unique/

This talks about how the difference between ape and human brains is more about the epigenetic timing of cell division and growth schedules than the expression of any particles coded proteins. — apokrisis

:up: I don't understand one thing: In the article you gave a link to, it says junk DNA are parsitic but DNA replication is supposed to be high precision machinery - like the best of atomic clocks - and any malfunction should manifest itself in catastrophic failure in an organism e.g. viral DNA being inserted into host DNA causing disease and point mutations causing illnesses like sickle cell anemia. It's like an organization that depended on atomic clocks to do businesd failing to notice that the precision time piece they have is giving the wrong time.

My personal view is that junk DNA, although not carrying information for proteins, serve a purpose as punctuation marks a la full stops, commas, etc. in the language of DNA.

but DNA replication is supposed to be high precision machinery - like the best of atomic clocks - and any malfunction should manifest itself in catastrophic failure in an organism — TheMadFool

Not really. Genes can jump about and still function. The genome parasitising itself is a neat extension of Darwinian competition that Crick himself suggested.

So biology isn't actually machinery even if it produces some remarkable machinery.

Check this and wonder how something so nifty could have evolved....

My personal view is that junk DNA, although not carrying information for proteins, serve a purpose... — TheMadFool

Sure. The article says its a mix of actual junk (all the parasitic transposons) and epigenetic machinery. I think the cited split was 45/55.

Check this and wonder how something so nifty could have evolved.... — apokrisis

I think a lot of people assume that DNA evolved - but how could natural selection apply before inheritance, and how could inheritance exist before something to code it?

Amazing video is all I can say.

I did some research of my own and it seems one geneticist, a Sydney Brenner (1927 - 2019) had a fondness for making the distinction between junk and garbage with reference to DNA, saying that junk is something you keep and garbage is something you throw out. He, if I'm correct, was of the opinion that "junk" DNA had some purpose and that it isn't useless garbage.

I think a lot of people assume that DNA evolved - but how could natural selection apply before inheritance, and how could inheritance exist before something to code it? — Wayfarer

First, organic chemistry then, DNA and only then, natural selection. I think that's how it played out anyway.