Why? Support the contention of unreasonable, and explain the colossal variations in genome size between even very similar species. This isn’t just about humans.
Does the Fork fern really need a 160gbp genome? Why do some water fleas the size of a couple of mm have 10gbp genomes?
Why do genome sizes vary by a factor of 5 between different species of onions? Why is it larger than the oh so complex and special human genome for many of them?
Why does the vast majority in these variations come from poorly conserved dead transposable elements?
Absolutely. At least, in so far as he’s outputting nonsense, not that he’s spamming this forum with discovery institute articles. Are Nobel laureates never wrong?
It’s not about lncRNAs but any sort of transcription.
Larry Moran writes about it here:
About half the human genome is expressed as introns at the level of the associated functional protein coding genes. Those introns are still not conserved, contain many dead transposable elements, and are typically quickly degraded again.
More equivocation on the word “neutral”. There are no strictly neutral mutations. Take up your disagreement with Eyre-Walker and Keightley. The ratio of deleterious-to-beneficial (neutrals included) is 1000:1 at best.
You say “as background fitness goes down”, implying the exact opposite of what is actually observed. The effect you’re citing causes fitness to stop increasing. It does not cause fitness to move in the opposite direction after a decline and suddenly start increasing.
Again, you’re inversing reality. The experiments you’re appealing to are 1) microbes (not LMEs) and 2) they are acclimating to lab conditions. Fitness begins to increase and then stops increasing, just like the name “diminishing returns” clearly indicates. This has absolutely nothing to do with rescuing a population from GE once fitness decline has set in! Just ask the Wooly Mammoth.
And again, what is the underlying mechanism that you’re suggesting should cause this effect? This is exactly like saying that as you continue to make typos in an instruction manual, and the manual gets worse and worse, then your typos should gradually start to improve the manual instead of hurting it. In other words, nonsense.
Still, personal incredulity isn’t an argument, is it?
That’s certainly the central argument, though there are others. And again you have nothing more than personal incredulity.
I don’t think that’s true. Let’s remember that only about 20% of conserved DNA is protein-coding.
What is this nature that gives you any reason to suppose so? These hypothetical functions seem to survive complete deletion of the presumed relevant sequence. How does that work?
What do you think about fern and salamander genomes? What about the fugu genome? And are you familiar with the onion test?
That’s a caricature position that nobody ever held. “We don’t know its function” has never been a reason to suppose that a sequence is junk. Are they conserved? Well, the functional ones are.
They exist in such concentrations in the cells in which they were detected. You could always hypothesize that we haven’t look in the right place, but is that a good reason to assume that the whole genome is functional?
They’re not exactly observations, are they? They’re predictions from an AI model. But I don’t know what to make of it. Note, however, that this transcription is apparently sequence-dependent, which suggests that if it were functional it ought to be conserved. And note that the authors suggest a few non-functional explanations.
If deletion tolerance proves non-function in non-coding sequences, it should prove the same for protein-coding genes. Yet, according to the abstract of the paper below, 60% of yeast genes can be individually deleted without obvious phenotype under laboratory conditions. Does this mean that 60% of the yeast proteome is junk?
I think the fugu example is working against you. Indeed, it happens that human and fugu have approximately the same number of protein-coding genes. Yet their biological complexity differs very significantly, with human more complex than fugu. That complexity difference must be encoded somewhere. And the most parsimonious explanation is that it resides in the non-coding regulatory architecture, ie., the part of the genome you dismiss as junk. Fugu’s compact genome tells us what a minimal vertebrate regulatory system looks like. The human genome’s additional complexity is not excess; it is the genomic signature of what separates us from a fish.
As to why some less complex creatures than humans have nonetheless larger genomes (salamander, ferns, oignon etc…), it is an open question but that doesn’t by itself vindicate the junk DNA hypothesis, for “we cannot explain this” is not the same as “it has no function.”
No proof, just evidence. But “obvious phenotype” is an extremely weak assay for selection, and “laboratory conditions” could be a problem too. I will again suggest that conservation is the best evidence.
Could you try responding to the rest of my questions?
Just to add some perspective as to why this is a perfectly reasonable position:
Suppose that each and every cell in a human body - all 30 trillion of them - is specified by a unique, singular combination of transcription factors. How many such factors would be required to uniquely specify each and every cell? Not tissue, not organ, but every single cell?
The answer is 45.
What does this simple thought experiment show us? It shows us that one does not need an incredibly, impossibly large amount of genetic information to generate all of the cells, cell types, tissues, etc. that we see in the human body. The notion that 10% of the human genome cannot possibly specify all this incredible complexity is wrong - just plain wrong. 300 million bp of information is easily up to the challenge. And then some.
Right. So you have no explanation for the data, just the assumption that it’s functional. This is exactly the advantage of the junk DNA hypothesis, that it explains all the relevant data.
There is a mechanism that explains why it varies so much between species and even individuals within species (it’s mostly selfish transposable elements and repetitive DNA).
That also explains their sequences (they are, in fact, derived from transposons, and the sorts of deletions and duplications that occur easily in repetitive DNA).
It explain why they’re not conserved (because it’s mostly nonfunctional).
It explains why it’s mostly silenced (again, because it’s transposons and they have regulatory elements within them which, if active, alters expression of nearby genes). And so on and so forth.
Notably this relatively simple explanation (being simple is one of the three major hallmarks of a good scientific theory) explains almost all the relevant data in all known species (explaining a lot of data is the second of the three hallmarks of a good scientific theory), and finally it is trivial to falsify it for any given locus: Demonstrate effect on a trait/phenotype and the junk hypothesis is instantly falsified (the third major hallmark of a good scientific theory, it should make easily testable predictions). Junk DNA checks all three boxes.
In contrast, the “it’s almost all functional even if we don’t presently know how” is in all respects a worse scientific hypothesis. It presently explains none of the relevant facts, it simply assumes that this will be done in the future. Every species has it’s own unique “ahh but here it functions in some new and different way” rationalization. And it’s almost impossible to falsify because you can just keep arguing that no matter what we just don’t know enough to really say it’s nonfunctional.
The possibility of having a typo where there has already been a typo, such that the 2nd typo in the same place (i) doesn’t make the manual worse, and (ii) may even make the manual better by reversing the first typo, is clearly beyond some-one.
You think well of yourself. Where lies this greater complexity, and it is reasonable to suppose that several times the amount of non-coding DNA is requlired to make a human than to make a fugu? And are you familiar with the concept of the dog’s-ass plot?
Nobody ever claimed that. The point is that genome size is not correlated with complexity, by whatever measure you claim. Thus the c-value paradox. And that suggests that most DNA in large genomes doesn’t do much, if other species of similar complexity can do without it. You keep appealing to ignorance when I appeal to data.