I didn’t claim they identified a few functional members of a particular class of Junk DNA and concluded that the entire class must therefore be functional. I never said anything like that. That’s NOT my criticism… not even remotely. See next for my ACTUAL critique:
Wrong! Completely wrong. Once again… the claim that 80.4% of the human genome is ‘functional’ is based on their own definition of function; which is so broad such that just ONE instance of biochemical activity (e.g. RNA transcription) in just ONE cell type is sufficient to meet their personal criteria for ‘functional’. And no, this is not misrepresenting ENCODE. This reason behind the claim that 80.4% of the genome is functional is plainly spelled out in their paper. Here is the line again:
An integrated encyclopedia of DNA elements in the human genome | Nature
The vast majority (80.4%) of the human genome participates in at least one biochemical RNA- and/or chromatin-associated event in at least one cell type. Much of the genome lies close to a regulatory event: 95% of the genome lies within 8 kilobases (kb) of a DNA–protein interaction (as assayed by bound ChIP-seq motifs or DNase I footprints), and 99% is within 1.7 kb of at least one of the biochemical events measured by ENCODE.
So, NO! They did NOT (as you mistakenly claimed) make any effort to ensure that the activity they measured was due to function. They just assumed that any activity they detected = ‘function’. No effort was made to distinguish signal from noise in the data they collected. That’s the very issue that ENCODE was criticized for, which I am echoing here (to your deaf ears).
It’s not the ‘explanation’ that’s the issue. Although, ENCODE doesn’t really provide an explanation, not even a hypothesis for the data they collected. Still, that’s not the issue. The problem is their reasoning that relies on a tautology. They simply have their own definition of ‘functional’ wherein any biochemical activity of at least one event in at least one cell type = ‘functional’… SOUND = SIGNAL… A definition of function that doesn’t consider noise, not even as a remote possibility. According to this line of thinking, non-functional biochemical activity due to noise is just as logically impossible as a married bachelor. The reality of noise in biology (e.g. spurious transcription) alone is enough to demonstrate how inappropriately loose their definition for ‘function’ is.
Here is the FULL section for context, including the key sentences that you conveniently left out (highlighted with bold letters):
5.3. Isochores and Non-Coding DNA
The discovery of non-coding DNA (meaning here DNA that does not code for proteins) goes back to the time when the number of genes, which, up to the late 1960s, was thought to be 1 million and to correspond to the totality of the human genome, started to decrease with the discovery (thanks to hydroxyapatite chromatography) of repeated sequences. [94] It has now reached a new low of 21 306,[95] a mere ≈2% of the human genome.
Three main explanations were put forward for the existence of non-coding DNA, a problem deserving of the term “mystery,” given that it had withstood 50 years of probing. Ohno, [89] mostly focusing on pseudo-genes, proposed that non-coding DNA was “junk DNA.”[96,97] Doolittle and Sapienza[98] and Orgel and Crick [99] suggested the idea of “selfish DNA,” mainly involving transposons visualized as molecular parasites rather than having an adaptive function for their hosts. In contrast, the ENCODE project[100] claimed that the majority (≈80%) of the genome participated “in at least one biochemical RNA- and/or chromatin-associated event in at least one cell type.” This claim, however, was rejected, mainly because of the loose definition of “functional” elements, in favor of the view that “junk DNA” or “selfish DNA” correspond to 80–95% of the human genome. [101,102]
At first sight, the pervasive involvement of isochores in the formation of chromatin domains and spatial compartments seems to leave little or no room for “junk” or “selfish” DNA. HOWEVER [capitalization emphasis added],one should now consider that coding sequences are compositionally correlated with the isochores in which they are located (the compositional constraints of the latter depending upon the need to encode/mold chromatin structures as already mentioned) and yet they are expressed. This indicates that there is no problem for transposons to be, on the one hand, compositionally correlated with the “host” isochores, and on the other, to be active. [103] Needless to say, this view also leads to an understanding of the “overlapping” transcription of long non-coding RNAs that originate from the majority of DNA sequences and that plays important roles. [104]
In the first line that you cut out, Bernardi mentions the actual criticism against ENCODE. The exact one regarding ENCODE’s definition of ‘function’ that I have repeatedly explained to you here. The fact that you must have read this line and consciously decided to snip it out of the quote is extremely telling. It indicates that you aren’t missing my point simply due to ignorance. You are deliberately trying to avoid the critique of ENCODE like the plague, even to such an the extend where you will purposefully delete any lines that mentions it within sections of your citations that you wish to quote.
This is similarly the case for the second part that you cut out. The last paragraph includes two integrally related parts:
- The first part of the paragraph, which notably starts with the phrase “At first sight…”, mentions how the pervasiveness of isochores seems to leave little room for ‘junk’ or ‘selfish’ DNA.
- The very next part, notably starting with the sentence adverb "However…, explains why the implication mentioned in the previous sentence may actually NOT be the case.
You CUT that second part out, thereby removing the relevant context for the first sentence, in order to make quote appear to align more with your preferred narrative. There is more to be said about isochores, however…