More Recent Comments

Saturday, October 13, 2018

The great junk DNA debate


I've been talking to philosophers lately about the true state of the junk DNA controversy. I imagine what it would be like to stage a great debate on the topic. It's easy to come up with names for the pro-junk side; Dan Graur, Ford Doolittle, Sean Eddy, Ryan Gregory etc. It's hard to think of any experts who could defend the idea that most of our genome is functional. The only scientist I can think of who would accept such a challenge is John Mattick but let's imagine that he could find three others to join him in the great debate.

I claim that the debate would be a rout for the pro-junk side. The data and the theories are all on the side of those who would argue that 90% of our genome is junk. I don't think the functionalists could possibly defend the idea that most of our genome is functional. What do you think?

Assuming that I'm right, why is it that the average scientist doesn't know this? Why do they still believe there's a good case for function when none of the arguments stand up to close scrutiny? And why are philosophers not conveying the true state of the controversy to their readers? I'm told that anti-junk philosophers like Evelyn Fox Keller are held in high regard even though her arguments are easy to refute [When philosophers talk about genomes]. I'm told that John Mattick is highly respected in philosophy circles even though knowledgeable scientists have little use for his writings.

Can readers help me identify papers by philosophers of science that come down on the side of junk DNA and conclude that experts like Graur, Doolittle, etc are almost certainly correct?


Image Credit: The cartoon is by Tom Gauld and it was published online at the The New York Times Magazine website. I hope they will consider it fair use on an educational blog. See: Junk DNA comments in the New York Times Magazine.

27 comments :

Faizal Ali said...

Why do philosophers even care about this subject, or feel they have anything to contribute to the debate?

T Ryan Gregory said...

Does it count if they're papers Ford and/or I have co-authored with philosophers?

Steve Greene said...

Philosopher of science discussing the field of biology - and especially genetics - certainly should care about the subject. In regard to contributing, one of the explicit tasks of a good philosopher is to contribute to the clarifications of concepts. Additionally, a good philosopher working on a particular topic should be able to articulate the various concepts that scientists working on the subject are employing, and be able to translate varying language references into a common terminology so to reduce confusion. (Of course, this is part of conceptual clarification.)

In regard to dealing with empirical claims, then a good philosopher - who, as not being a scientist, is not actually doing empirical studies and experiments himself - should be seeking to understand and articulate various positions and why scientists are making differing claims and arguments, and at least making sure that what the scientists are saying is well understand in a common conceptual terminology - even if the philosopher doesn't happen to take a side in a debate.

We don't need to denigrate philosophers doing philosophy - we just need to make sure they're doing their jobs adequately.

Larry Moran said...

Partly, but I’d to find something written by philosophers, for philosophers. Are there papers in the philosophical literature that challenge the views of Fox Keller, for example?

Joe Felsenstein said...

I cannot suggest philosophers who satisfy Larry's requirements, but I can suggest answers to his other question:

Assuming that I'm right, why is it that the average scientist doesn't know this? Why do they still believe there's a good case for function when none of the arguments stand up to close scrutiny?

1. In 2012, they heard that junk DNA had been disproven by a consortium of 400 genomicists. The official publicity machine of that consortium was insistent on the fact, and many science journalists echoed that. The scientists haven't heard anything nearly as well-publicized since, so they assume that's the last word.

2. They (most cell and molecular biologists and many genomicists) are used to thinking of the cell as a finely tuned machine, every aspect of which is there for a reason.

3. When they were young, they heard many statements by evolutionary biologists such as Ernst Mayr and Theodosius Dobzhansky about how perfect adaptation was and how every aspect of the phenotype was optimal.

4. They are not used to population geneticists' and molecular evolutionists' way of thinking, that the genome is the outcome of many processes, not all of which are natural selection, and even natural selection can imperfectly reflect current environments.

5. As a result of that, they have no feel for how small selection coefficients can be impotent in the face of genetic drift. (Which is a bit strange since many of the same genomicists also regard a selection coefficient of 0.001 as "weak selection" which probably has no effect).

6. Last but not least, if all that DNA is doing something think of all the grants that can be gotten to work that out!. A whole lifetime of work!

S Johnson said...

Looking from the outside, it seems to me the likeliest reason is that junk DNA isn't adaptationist in the usual sense of omnipotent natural selection. On the other hand adaptationism is natural theology without tedious denominational issues.

Most philosophers reject atheism, the position that we can say there is no God (magic) in favor of naturalism. Insofar as this means something, t appears to mean things like "science doesn't provide definitive knowledge about the real world" or "empirical demonstrations on ontological propositions are logically incoherent" or "scientific claims are inherently limited and never conclusive."

W. Benson said...

Another point is that many people probably think that a genome that is 80%-90% non-functional would result in large, unacceptably wasteful expenditures of energy and scarce nutrients, such as P and N, making high levels of "junk" unlikely to evolve through 'mere' drift. This at least was one of the considerations I had to resolve before becoming completely at ease, thanks to Sandwalk, with junk DNA. It is unfortunate that molecular biologists, people who have access to the data, have not provided information on metabolic and material cost (wastage) of junk DNA, as a percentage of total cell needs. I think this value, if as small as I suspect, would help convince many that junk DNA can be practically ignored as a drain on cellular activity and a factor reducing Darwinian fitness. Quoting Dan Graur (2018 - you can Google the source), "The assumption that a vast fraction of the genome evolves in a neutral fashion means that this DNA does not tax the metabolic system of eukaryotes to any great extent, and that the cost (e.g., in energy, time, and nutrients) of maintaining and replicating large amounts of nonfunctional DNA is negligible." To many people this is counter intuitive. Moreover, I am not aware that Graur's affirmation has ever been shown true or brought forward as evidence in the junk-DNA controversy.

Jmac said...

I have always been under an impression that the Onion Test has not been used as an argument for Junk DNA. What's changed?

Jmac said...

The phenomenon of genome sizes between relatively simple organisms with large genomes and complex ones with smaller genomes clearly indicates that complexity of many organisms doesn't exclusively depend on DNA. The cell differentiation processes gives us just one clue. The information needed for cell differentiation can't come exclusively from DNA as per category theory which stems from the set theory.

John Harshman said...

Nothing has changed. Your impression is wrong. The Onion Test has always been used as an argument for junk DNA, and in fact that's what it was invented for.

John Harshman said...

Your little syllogism only works if most of the genome is functional. But if it's junk, then there is no supposed relationship between genome size and organismal complexity for you to explain by your alternative.

The Other Jim said...

Where is that cartoon from?

Marcoli said...

Nick Lane had made the argument that eukaryotes experience only a slight energy burden from having a large # of genes because they have mitochondria (https://www.nature.com/articles/nature09486). Mitochondria are expandable in number and can increase cristae membrane surface area, resulting in a surfeit of ATP. This relieves the energy cost of having many genes. One supposes this could also permit the upward drift of having lots of junk DNA.
This has been found unnecessary, since large cell size alone is found to be sufficient to reduce the cost of having a large gene number (http://www.pnas.org/content/early/2015/10/29/1514974112). Again, this too could permit the upward drift of junk DNA.
I am not sure if this is wrong, but it seems to me that both views may be relevant. Mitochondria could permit eukaryote cells to evolve to large size, under directional selection, and then large size would permit large and largely junk genomes by drift.

Robert Byers said...

I think its meant the scientist who has these subjects as thier study. The others only know their own stuff.
I think also error is common in subjects of science because mostly they accept the established conclusions. They don't keep up with criticisms or that conclusions are no longer conclusive by the scientists in these subjects.
I think it moves in small circles and not reading updates.
As a creationist i find/found a lack of competence in all sciences that i bump into.
thats why its so easy for YEC/ID to put in doubt in substantive ways origin conclusions.
Right or wrong it comes down mostly to acceptance of authority.
Only a little weighing the evidence even for those in the subject under investigation.

br56u7 said...

Shouldn't the co author of mattick 2013, Marcel dinger, be included along with mattick?

Zencycle said...


Mr. Moran,
I'm not a biologist of any flavor, but have recently been sucked into the drama over at talk.origins. One contributor posted a link to your recent 'junk dna debate challenge'. I was doing some reading on the junk dna issue and came across this paper:

"A conserved function for pericentromeric satellite DNA"

https://elifesciences.org/articles/34122

"we propose that satellite DNA is a critical constituent of eukaryotic chromosomes to ensure encapsulation of all chromosomes in interphase nucleus. Our results may also explain why the sequences of pericentromeric satellite DNA are so divergent among closely related species, a contributing factor that led to their dismissal as junk."

Science daily gave a review that essentially states:

' the findings indicate that pericentromeric satellite DNA [aka 'junk dna'] performs the vital function of ensuring that chromosomes bundle correctly inside the cell's nucleus, which is necessary for cell survival. And this function appears to be conserved across many species.'

I'd be interested in your take on it.

Georgi Marinov said...

And what about species with holocentric chromosomes (e.g. C. elegans and a number of others)?

Larry Moran said...

Here's my take from Chapter 2 of my book in progress.

"Centromeres

Centromeres are the part of a chromosome required for proper segregation after the DNA has been copied and the cells are dividing (reviewed in Bloom, 2014). Each one of the pair of newly replicated chromosomes has to be separated and segregated into one of the daughter cells. The centromeres in chromosomes are the places where the spindle fibers attach in order to pull the chromosomes to opposite ends of the dividing cell.

We’ve known about centromeres for 50 years but it wasn’t until the 1970s that the DNA content of these regions was clarified. The centromere regions consist largely of highly repetitive DNA sequences. There are several different types but most of them are sequences called alpha-satellite sequences (Schueler et al., 2001). They are usually represented as gaps in the published human genome sequence because they consist of thousand of copies of short repeats of 100-200 bp (Malik and Henikoff, 2002). Centromeres are part of the heterochromatin in normal non-dividing cells.

There are 24 different centromeres in humans and the average size is about four million base pairs. Centromeres make up about 3% of our genome and most of that sequence is not junk although some of it can be deleted without doing any serious harm. This is a well-known part of the functional noncoding region of the genome. Let’s assume that the truly essential fraction of centromere sequences makes up about 1% of the genome and the other 2% is dispensible. This is a pretty good estimate.
"

I don't know of anyone who ever claimed that all centromere sequences are junk.

Larry Moran said...

check out ...

What's in your genome?

Divalent said...

One possible problem with touting the onion as an example of "big genome - no problem" is that it's a domesticated species. And so, like most domesticated species, has been selected for traits that are useful to human in a human-defined environment, which in many cases make them less fit if they were forced to compete in the wild without human assistance.

If a desirable trait means a human will ensure that all 50 of your offspring survive into the next generation, that can compensate for a pretty large fitness penalty that would otherwise manifest itself the wild.

Are there similar examples of big genomes in species (related or not to the onion) that weren't ever domesticated? If so, I think they would be a better example.

[My apologies if this comment has been double (or triple) posted: my Chrome and Firefox browsers are both having difficulty loading this site.]

Zencycle said...

I don't think they were stating _all_ centromeres are junk, in fact they limit it exclusively to pericentromeric satellite DNA, and they do it at least twice (that I could find, I don't have vast access to research libraries); in the article above and here:
"Function of Junk: Pericentromeric Satellite DNA in Chromosome Maintenance" http://symposium.cshlp.org/content/82/319.full

To narrow the question a bit more - would you consider that pericentromeric satellite DNA is (or was) classified as junk?

John Harshman said...

Perhaps you would prefer the fugu test. Ask yourself why you need a genome 8 times the size of a fugu genome. Or do you consider humans to be domesticated? Then how about chimps?

Jonathan Badger said...

I think that a "fugu test" isn't going to be very convincing to people who believe in human exceptionalism, because it is easy to convince oneself that humans are way more complicated than a fish and all that "junk" in the human genome is responsible. These tests work best when comparing very similar organisms with different genome sizes. Then it is obvious that the larger genome must be full of junk.

John Harshman said...

True, but then again one can always reject the onion test because "they're just onions", with no relevance to human beings. There is no way to convince the unconvinceable.

Larry Moran said...

I don't think you understand the Onion Test. It's a kind of thought experiment to challenge people who think they have come up with a good explanation for all that extra DNA. If you think it's required for regulation then why do some species of onion require five times more DNA and why do other onion species require 4 times more than that?

The onion test

The only reasonable explanation is that all the extra DNA is required in humans but the excess DNA in onions is junk. The fact that some species (like fugu) have smaller genomes must mean they are less complex, according to this view. It follows that humans are unique in that they have achieved the perfect genome full of function—not too big and not too small.

This is the "Goldilocks genome" (Doolittle, 2013; Doolittle and Brunet, 2017).

c0nc0rdance said...

I've always been on the non-functionalist side of this debate, although I'm more in the applied side of biology (diagnostics), but the strong evidence that lncRNA has utility as a biomarker for disease states gives me a certain pause. Do you have any thoughts on how that interacts with assigning function to junk DNA?

Larry Moran said...

I’m not sure what you’re referring to. I can think of two possibilities.

1. One of the functional lncRNAs has something to do with disease. This has nothing to do with junk DNA.

2. One or more spurious transcripts that are long and noncoding are associated with with some disease state. This probably happens because the disease alters the normal state of gene regulation. The spurious transcripts could come from regions of junk DNA.