If the spaces between genes are important then mutations that change those spaces should be harmful part of the time and be selected against.
Evidently and crystal clear, but:
We do see evidence of negative selection against indels, but only in 10% of the genome.
Do you mean the 10% functional part of the genome affected by those indels, or, to be precise: do you mean that 10% of the indels have functional effects?
Hence:
If that were the case throughout the genome, shouldn't we see selection against indels in 100% of the genome?
Why should it? (again the question!)
Because 85-88% (after ENCODE) of the genome is merely junk.
When one of those parts is deleted by an indel, nothing happens with selective effects.
When something is inserted there, nothing happens as well unless the very rare instance this newly inserted chunk is immediately functional. Such insertions might get functional only after more future mutations. But measured just after the moment of insertion, no selection will be detected.
My contention was:
Denisova writes:
An indel that impairs the spacer would be detected by the adjacent genes not being transcribed properly.
Now why would we need to see indels having selective effects in 100% of the genome when such effects are only to be expected in the 12-15% functional part of the genome? Moreover, indels may have direct effects too. If an indel deletes parts of the gene that encodes for cytC, for instance, it's over and out and a very strong selective effect takes place. Now, I wasn't referring to these direct indels. I was referring to indels that ONLY hit the (rather tiny) parts of the spacer DNA that separate genes.
To be precise: two particular genes may be separated by larger chunks of spacer DNA. Hence, my argument only counts for the instances where genes are separated by rather
small parts of spacer DNA. The larger such a spacer chunk, the less the odds of it completely deleted by an indel.
So, I do not understand your argument "shouldn't we see selection against indels in 100% of the genome?"
Denisova writes:
I think that accounts for notable amounts of energy consumption. And notable it is when a reduction of this consumption, without any functional consequences, would retain energy for other purposes.
Both of those are empty assertions.
Well, I said I do not know any empirical studies on the energy budget of cells in relation to the energy consumption of transcription - maybe there are some approaches but I do not know any of those - but I tried to make my case by arguing why this energy consumption should be notable.
Hence, those arguments were not assertions but arguments.
And their emptiness is still to be substantiated by you!
Denisova writes:
Why do you call it "low level" transcription?
In this context, low level transcription refers to the number of copies of any given RNA transcript. For the junk DNA that ENCODE found mRNA copies for, the number of copies were several orders of magnitude lower than commonly transcribed genes such as B-actin. In fact, these areas had so few copies that they had to use special techniques to detect them.
If you want to make the argument that the observed rate of junk DNA transcription is energetically expensive then you need to have a handle on the number of RNA copies and bases per copy since the energy is spent making the nucleotides and linking them with phosphates.
OK, I understand now what you meant with low level transcription.
But as I understood, the making of nucleotides does energetically not come very cheap.
What you say is that junk DNA tends to be low level transcribed.
The low level transcription is already incorporated in my arguments. The tendency of junk DNA to be low level transcribed is counteracted by its majority in the genome. The 76% of total DNA transcribed minus the 12-15% found to be functional. Even when, as ENCODE also found, 70% of the transcription coverage was less than 1 transcript per cell.
What matters is the amount of energy that is expended transcribing junk DNA. If it is just 1% of the energy compared to transcribing functional genes then it isn't a problem.
I agree but I was curious what
precise share in the total energy budget it exactly is taking away.
When comparing orthologous cytC genes between species these would show up as regions of conservation. When those are found within an open reading frame, it indicates function. Junk DNA has none of these features. It doesn't have regions that are highly conserved between species. It doesn't have open reading frames that have upstream ribosome binding sites, for example.
Of course junk DNA doesn't have such features. Otherwise, like cytC, it would be functional.
The functional parts of cytC are extremely conserved, otherwise, as I mentioned, it wouldn't be possible to transplant human cytC to algae without doing the latter any harm. But I was referring to the
redundant parts of cytC. The 70% of its amino acids that seem to only be freeloading.
Edited by Denisova, : No reason given.
™ Version 4.2
(1)
