Junk DNA, Rana, and Cardinale

[swamidass (S. Joshua Swamidass)]

@dsterncardinale interviews Fuz from RTB in Junk DNA. Seems like this might turn out to be a productive conversation.

Junk DNA! with Dr. Fazale Rana

[dsterncardinale]

For what it’s worth, I thought this was really fun and very productive. But I’m biased.

[John_Harshman]

Just for purposes of discussion: productive in what way?

[Art (Arthur Hunt)]

One thing that people like Dr. Rama need to be pushed on - many (most) of the functions they see in non- protein coding DNA are, by ID standards, low or zero information functions. Structural roles for (organizing, nuclear volume, etc.), microRNA targeting, binding of individual transcription factors, these and more involve short and/or minimally conserved sequences. This flies in the face of the ID party line.

If ID is about information, then what we know about the functioning of noncoding DNA is not consistent with ID theory.

[colewd (Bill Cole)]

Hi Dan
I sent this discussion to a couple of friends and called you Dr Dave by mistake again. I did send a correction.

Congratulations on this video. I think it is an example of two different scientific views finding common ground and learning from each other.

One aspect I would consider is the measurement needs to include level of transcriptional activity during embryo development.

[Art (Arthur Hunt)]

Along the lines of junk (in general), it bears mentioning that cells devote a significant proportion of their genomic budget to turnover and degradation. Thousands of genes encode a machinery that throws aways much (most) of what the cell makes (nucleic acids and proteins). Among other things, these surveillance systems target the products of transcription of junk DNA. Thus, if one is compelled to infer function from transcriptional activity, one implies that the function of an untoward fraction of the eukaryotic genome is to make, for want of a better term, garbage.

Chemically and thermodynamically, this all makes perfect sense. But from an ID perspective, it is ludicrous.

[dsterncardinale]

@John_Harshman

Just for purposes of discussion: productive in what way?

Productive in that we were able to drill down into the specifics of where we agree and disagree and why that’s the case for the specific places where we disagree. I feel like a lot of these discussions is talking past each other, so I always try to get everyone involved to clearly and specifically articulate what their actual argument is before getting to the “disagree on the internet” part, and I think this is a pretty successful example of that.

[colewd (Bill Cole)]

@Art

Chemically and thermodynamically, this all makes perfect sense. But from an ID perspective, it is ludicrous.

Hi Art
Are you referring to the ubiquitin system?

[Mercer (John Mercer)]

Obviously not. While some people claim that there are no stupid questions…

[Giltil (Gilbert Thill)]

@Art

Along the lines of junk (in general), it bears mentioning that cells devote a significant proportion of their genomic budget to turnover and degradation. Thousands of genes encode a machinery that throws aways much (most) of what the cell makes (nucleic acids and proteins). Among other things, these surveillance systems target the products of transcription of junk DNA. Thus, if one is compelled to infer function from transcriptional activity, one implies that the function of an untoward fraction of the eukaryotic genome is to make, for want of a better term, garbage.

This is weird. In essence, what you are saying is that if one infers function from some phenomenon P giving rise to some entity E targeted by some surveillance system, then one implies that the function of a fraction of the genome is to make garbage. Would you say that for P=translation and E=polypeptides ?

[John_Harshman]

@dsterncardinale

Productive in that we were able to drill down into the specifics of where we agree and disagree and why that’s the case for the specific places where we disagree.

Fair enough. I may actually have to look at the video. One assumes that the specific places are intractable. True?

[swamidass (S. Joshua Swamidass)]

I thought one of your analogies, which Fuz seemed to get, was really helpful.

You discussed how drafty windows and rafters will make a room cooler, and the thermostat will need to compensate. The drafty features are contributing to the temperature, but it would be a mistake to think that their purpose is to lower the temp so the thermostat compensates.

That’s a good way of explaining on difference between biochemical activity and function, as ID means it.

[Art (Arthur Hunt)]

@Giltil

In essence, what you are saying is that if one infers function from some phenomenon P giving rise to some entity E targeted by some surveillance system, then one implies that the function of a fraction of the genome is to make garbage.

IF one were to infer function. The alternative is to accept that there is no function for said phenomenon. @swamidass mentioned an analogy described by @dsterncardinale:

@swamidass

You discussed how drafty windows and rafters will make a room cooler, and the thermostat will need to compensate. The drafty features are contributing to the temperature, but it would be a mistake to think that their purpose is to lower the temp so the thermostat compensates.

There is no need to infer a function for the draft. It exists owing to the features of the room. Likewise, for much (most) of the DNA in the eukaryotic genome, the features noted by ENCODE and other research that are associated with so-called junk DNA are like the draft - they do not reflect any sort of function but rather the inevitabilities associated with the biochemistry that underlies much of the workings of the cell.

@Giltil

Would you say that for P=translation and E=polypeptides ?

Generally speaking, no. But I am not sure what is meant by the question.

@colewd

Hi Art
Are you referring to the ubiquitin system?

That, as well as RNA surveillance systems (of which there are several).

And more. Consider, for example, the fact that plants possess two novel DNA-dependent RNA polymerases (Pol IV and Pol V). These two enzymes are sources of transcripts derived from particular genomic regions. These RNAs do have function - they work to suppress transcription of these regions by RNA polymerase II. Pol IV/Pol V target repetitive DNA (transposable elements and the like), and thus reduce or eliminate transcription of repetitive DNA.

Those who think along the lines of Mattick et al. would likely view the products of Pol IV/Pol V transcription (short RNAs with no protein-coding capacity) as evidence for function of repetitive DNA itself. However, they are reflections of rather the opposite.

[Roy (34-year veteran)]

@Mercer

Obviously not. While some people claim that there are no stupid questions…

There are no stupid questions… but there are stupid people.

[Rumraket (Mikkel R.)]

@Giltil

In essence, what you are saying is that if one infers function from some phenomenon P giving rise to some entity E targeted by some surveillance system, then one implies that the function of a fraction of the genome is to make garbage.

That’s what many anti-junk proponents (like the first press releases of the ENCODE fiasco) do. The mere fact that some locus is transcribed is inferred to make it functional.

They seem to completely ignore things like transcriptional level, degree of conservation, the actual properties of the sequence being expressed(for example, such as it being mutationally degrading fragments of ancient transposons), where in the body it is expressed, that much of it is actively suppressed, and crucially to Art’s point, how quickly it is degraded again and so on.

Given that all these factors are basically ignored and the mere detection of a transcript is taken as grounds for concluding it is functional, when we then learn that much of these junk transcripts are quickly degraded again, that at face value implies it’s function is to be degraded again.
It’s a reductio ad absurdum against these types of function inferences of course: We are not supposed to conclude that it’s function is to be degraded. Given an understanding of all the relevant data, the conclusion is it is functionless junk.

It is difficult to think of how else nonfunctional DNA could behave if it is insufficient to point out that it is being transcribed at a very low level, having sequences that are not being conserved by purifying selection, that look like degrading repetitive fragments of ancient transposons, which almost never leave the nucleus and have extremely short half-lives because it’s immediately targeted for degradation. Then what the hell does it take to convince anti-junkers that most of eukaryotic genomes are junk?

Their resistance to the junk conclusion from entirely reasonable and basic inferences from biochemical and comparative genetics-data is making their function-hypothesis basically unfalsifiable.

[T_aquaticus]

Larry Moran is (in)famous for his blog posts on junk DNA over at Sandwalk. One of my favorites over the last year or so is this article:

How many enhancers in the human genome?

The post focuses on this quote from the ENCODE paper:

The human genome contains only about 20,000 protein-coding genes, yet gene expression is controlled by around one million regulatory DNA elements called enhancers.

As laid out in the blog post, some quick math leads to a rather problematic conclusion. If these enhancers are all truly functional, then there are an average of about 50 enhancers per gene. No one knows of a single gene in the human genome whose expression is governed by 50 enhancers, much less 50 being an average. Just from this math alone, it becomes very apparent that not all of these enhancers are functional.

Moran also posted this quote:

… as a chemist, let me say that I don’t find the binding of DNA-binding proteins to random, non-functional stretches of DNA surprising at all. That hardly makes these stretches physiologically important. If evolution is messy, chemistry is equally messy. Molecules stick to many other molecules, and not every one of these interactions has to lead to a physiological event. DNA-binding proteins that are designed to bind to specific DNA sequences would be expected to have some affinity for non-specific sequences just by chance; a negatively charged group could interact with a positively charged one, an aromatic ring could insert between DNA base pairs and a greasy side chain might nestle into a pocket by displacing water molecules. It was a pity the authors of ENCODE decided to define biological functionality partly in terms of chemical interactions which may or may not be biologically relevant. (Jogalekar, 2012)

I think that perfectly sums up the entire debate. Chemistry is messy, and biology is even messier. It isn’t at all surprising that there is junk DNA in the human genome given the number of genes and the number of bases in the genome. We should expect non-functional DNA to be transcribed, and for random non-functional sequences to bind transcription factors. Add in the lack of sequence conservation, and the conclusion seems rather unavoidable.

[T_aquaticus]

@Art

If ID is about information, then what we know about the functioning of noncoding DNA is not consistent with ID theory.

That’s a point I made quite a few years ago:

If we accept the ID claim that these sections of DNA have function then it leads to the unavoidable conclusion that function is found in almost any DNA sequence. Function is not rare in DNA sequence space. Instead, function is found in almost every random DNA sequence, and no mutation can do away with that function.

As far as I can see, ID has to pick between “no junk DNA” and “function is rare”. They can’t have both.
No Junk DNA vs. Rarity of Function

Interestingly, Ann Gauger actually addressed these concepts over at ENV.

This question was posed recently at Peaceful Science, a discussion site that seeks to promote dialog between atheists, theistic evolutionists, and proponents of intelligent design. (Their success is mixed. ID proponents often feel like Aragorn in the last battle.) It’s a good question. It came out of a conversation about orphan genes, where I was arguing that non-coding DNA was extremely unlikely to give rise to a new coding sequence with any function. Yet ID people claim all the time that the majority of the genome is functional. How can sequence be both functional and non-functional at the same time? The answer turns on two things. The meaning of “function” and clarity about what’s being tested.
. . .

But being functional in the biochemical sense (à la ENCODE) does not mean it is easy to give rise to new genes and proteins. When we say functional sequence is rare in sequence space, we mean a different sort of function and sequence than in ENCODE. We mean a sequence that can have the ability to carry out an enzymatic reaction. It is our claim that proteins made from random sequence will rarely if ever have any sort of enzymatic activity.
In Arguments for Intelligent Design, Definitions and Assumptions Are Important | Evolution News

It would appear that @agauger chose “lots of junk DNA” for the human genome since only a tiny fraction of sequence in the human genome produces proteins with enzymatic function.

[Rumraket (Mikkel R.)]

Another very odd definition of function, since proteins can have structural, ligand, and other roles besides enzymatic activity.

[misterme987 (Andrew P.)]

And anyway, as we discussed in this thread, enzymatic activity isn’t particularly rare anyway (contra Axe).

[dsterncardinale]

Assuming you’re referring to Axe’s 2004 paper…I mean my goodness, the man just butchers the basics there. He “calculates” the probability of arriving at a specific functional protein fold and extrapolates that to the likelihood of any functional fold evolving. I mean, c’mon. That’s not even trying, even by DI hack standards.