More Recent Comments

Thursday, January 17, 2008

A Junk DNA Quiz

 
Take the junk DNA quiz in the left sidebar to let me know what you think of your genome. How much of it could be removed without affecting our species in any significant1 way in terms of viability and reproduction? Or even in terms of significant ability to evolve in the future? In other words, how much is junk?


1. I did not choose the word "significant" in order to be obtuse. I picked it in order to eliminate some trivial possibilities that really don't make any difference. For example, no matter how little DNA you delete you would be able to detect some change, even if it's just a reduction in the time to replicate the genome of the amount of energy used. If you think that such changes are "significant" then you should answer "none" to the question in the quiz.

22 comments :

T Ryan Gregory said...

I think this is where a distinction between nonfunctional and inconsequential is important. In terms of whether most DNA is functional, I would agree on the basis of what we currently know that much of it could be deleted in principle. However, this would also affect cell size, and therefore organs, and therefore organisms. And yet, I would not necessarily consider the influence of DNA amount on the cell as a function. It could very well be that there is upward pressure from transposable elements and other mechanisms that cause DNA to accumulate, and downward pressure against this accumulation through selection on organisms. The balance that has been reached is not necessarily adaptive for either side. However, deleting a lot of it could have impacts on development and morphology nonetheless. Maybe it would even be a beneficial change, maybe deleterious, but I can't assume that there would be no effect.

A. Vargas said...

While I agree that most DNA is non-functional and subtantial chunks of it can be deleted, I think it's not so easy to go ahead and say something like 90% in a single great deletion without ANY sigificant change.
The radically different time of DNA replication implied would probably have significant consequence on any mechanism regulating rates of cell proliferation.

Also: genomic, higher level chromosomal architecture is completely non-trivial, from the level of gene expression to the inheritance correlated traits (as chromosomes stick together, split apart, etc).
What would happen to chromosomal architecture if moe than 90% of the "bricks" suddenly missing? I'd expect some phenotype, to say the least!!
Further, from the interspecies correlation form genome size and cell size, we can expect that cells would be quite smaller with a 90% deletion. In itself, cell size could be counted as a significant effect.

The largest deletion without a phenotype known in mice is quite small, like 4%? And mice have plenty of junk.

Natural cases of genome reduction can give us a better idea of how much genome it is possible to lose, although in many cases it may not be possible to figure in how many steps did this reduction occur or whether there were or not immediate significant phenotypic effects.

For instance in the dinosaur-bird line, there has been a reduction in about 50% the size of the genome.

I guess the small genome of the fugu may be champion of all natural reductions, but I really don't know

Steve Reuland said...

Yeah, what he said. I voted 11-49% not because I think that at least 50% of the genome is functional, but because I doubt that you could really remove that much DNA all at once and not cause some kind of problem, at least from a structural standpoint if nothing else. The functional bits, particularly the complex regulatory apparatus, largely evolved within the context of huge amounts of non-functioning DNA, and removing it suddenly could well throw things out of whack.

Eric Pedersen said...

tr gregory:

I was thinking the same thing. Even if the sequence itself is functionally useless, I'm somewhat sceptical that deleting %75-%90 of the genome wouldn't throw a monkey wrench into the works. I'm still pretty sure that you could randomly alter the sequences of %80-%90 of the genome without really noticing anything at all.

Out of curiosity, does anyone know of any (good) attempts to analyze spatial distribution of junk DNA relative to functional sequences?

Anonymous said...

Are we talking simple deletion here, or something more sophisticated? By doing some clever engineering, I bet it would be possible to rearrange parts and circumvent many of the structural issues that would be a problem with just knocking large chunks out.

My guess is that somewhere in the neighborhood of 50% could be deleted, in a future where we have vastly more knowledge about the genome.

Larry Moran said...

sanders says,

The radically different time of DNA replication implied would probably have significant consequence on any mechanism regulating rates of cell proliferation.

We have plenty of examples of polyploid hybrids in plants where the amount of DNA doubled instantly from one generation to the next. There didn't seem to be any drastic consequences.

We have several examples of closely related species whose DNA content differs by 8 or 16-fold. You can't tell them apart unless you look at their nuclei.

Incidentally, in eukaryotes the time taken to replicate the genome isn't dependent on the amount of DNA.

T Ryan Gregory said...

We have plenty of examples of polyploid hybrids in plants where the amount of DNA doubled instantly from one generation to the next. There didn't seem to be any drastic consequences.

Not in mammals, though.

We have several examples of closely related species whose DNA content differs by 8 or 16-fold. You can't tell them apart unless you look at their nuclei.

Not in mammals.

Incidentally, in eukaryotes the time taken to replicate the genome isn't dependent on the amount of DNA.

Cell cycle time is, possibly due to the influence of nuclear size on cyclin accumulation.

T Ryan Gregory said...

I think a useful follow-up question might be, how much of the genome could you delete a small bit at a time without effect. So, let's say you had one set of subjects and deleted 1%. In a second set you deleted a different 1%. A third set, yet another different 1%. How many of those 1% bits could be taken out individually without effect? Maybe then you could say 50, 75, or even 90% could be deleted and my initial note wouldn't be applicable.

Anonymous said...

L.M.: We have plenty of examples of polyploid hybrids in plants where the amount of DNA doubled instantly from one generation to the next. There didn't seem to be any drastic consequences.
This is not relevant to the question posed for two reasons: 1) the question asks about reduction in genome size, not increase. 2) the distance between genes in polyploid species is unchanged.

A. Vargas said...

I know several cases of plants and animals where polyploidy indeed does have a significant phenotypic effect, specially on body size.
Plus I am pretty sure that I've heard about the meiosis in some salamanders taking forever becaase of their humongous aounts of junk dna.
Let's also not forget that cell size of specific tyoes of cells (for instance,bone) correlate nicely with genome size when compared across species (within the body, different cells of types have different sizes, of course)
How would this trend be so consistent if there were not increases and decreases of cell size accompanying genome size variation?

Stephen Matheson said...

I voted 11-49, not because I think that much of the genome has a "function" but because I'm not sure that outright deletion would be without consequence. (I'm echoing others here, I know.) A better (surely different) question might be: "How much of the human genome is merely taking up space?" which is a paraphrase of how Alex Palazzo put it. I think I'd still be in the 11-49 range, but I'd feel more strongly about it in that case.

I'm intrigued by the association of flight with smaller genome size, suggesting an example of TRG's "downward pressure," and pointing to a significant role for genome size (via cell size, it seems) in certain types of adaptation. Would love to be corrected on this if I'm off base here.

A. Vargas said...

Genome size reduction had already occurred in theropod dinosaurs, long before some of them evolved flight.
I'm not buying it. What about pterosaurs? They probably had larger, typically reptilian genomes. What's the genome size in bats, BTW?

SPARC said...

It's not clear to me if you mean how much of the genome one can delete and still have a vital organism (then I would vote for >50%). Or did you mean which parts of the genmome are principally dispensible if one would delete only one at a time. If you meant the later one could even delete quite some coding sequences without major phenotypic effects (therefore I voted 75-89%).

Sigmund said...

I presume Larry means how much of the human can hypothetically be deleted in one instance without major consequence. Deleting smaller amounts (such as the separate deletions of 1% in many separate experiments suggested by others on this thread) doesnt make sense in the context of the question - if that was the case why not reduce it down to a single base pair? Three billion experiments of a one base deletion and I suspect that you will get the result that you are around the 98-99% safely deleted. Taking it as a single experiment and I suspect the ultimate answer is in the 11-49% range. Like others here I suspect a lot of functional elements work in a location specific context (i.e. twenty kb away from another functional element - even though the 20kb is made up of non functional repeats) Deleting these types of repeats in this context would, however, affect the actual functional element despite the fact that it is only a tiny percentage of the overall sequence.

T Ryan Gregory said...

Genome size reduction had already occurred in theropod dinosaurs, long before some of them evolved flight.
I'm not buying it. What about pterosaurs? They probably had larger, typically reptilian genomes. What's the genome size in bats, BTW?


I believe a paper about pterosaurs is in the works, and they have small genomes (but I won't say it more conclusively than that as I have not seen the data). Bats have some of the smallest genomes among mammals. And, to clarify, theropods had smaller genomes than other dinossaurs but not as small as most modern birds. Genomes shrank more after flight evolved, especially in strong flying birds.

I'm intrigued by the association of flight with smaller genome size, suggesting an example of TRG's "downward pressure," and pointing to a significant role for genome size (via cell size, it seems) in certain types of adaptation.

Then let me recommend:

Gregory, T.R. (2005). The Evolution of the Genome. Elsevier, San Diego. (Chapters 1 and 2)

T Ryan Gregory said...

I presume Larry means how much of the human can hypothetically be deleted in one instance without major consequence. Deleting smaller amounts (such as the separate deletions of 1% in many separate experiments suggested by others on this thread) doesnt make sense in the context of the question - if that was the case why not reduce it down to a single base pair? Three billion experiments of a one base deletion and I suspect that you will get the result that you are around the 98-99% safely deleted.

Since it is just hypothetical anyway, I would be fine with trying it all at once and 1% at a time and 1 bp at a time to compare these.

Anonymous said...

Is variation in genome size measured within a species for any geographically widely distributed mammal? (Or are all the genome size data done at a higher taxonomic level, with a few measurements per species?)

If so, is there a link between genome size and body size that would implicate genome size as a factor in determining an adaptive cline of body size? (for instance, in a system where latitude correlates with body size, such as in lions or canids?)

This would be a test of the hypothesis that genome size is in some sens adaptive or related to an adaptation.

If this were the case, then I would vote for a "percentage deletion of junk" that would be limited by this effect. So, you can delete more of it if you are also willing to transport the resulting animals to an appropriate habitat where metabolic rate, body size, or some other feature would match the new form (I mean, there are ethical considerations here....)

Oh, by the way, Larry, thanks for starting up a discussion of this issue! Well done.

T Ryan Gregory said...

Is variation in genome size measured within a species for any geographically widely distributed mammal? (Or are all the genome size data done at a higher taxonomic level, with a few measurements per species?)

It is usually considered fairly constant for a particular species (hence "C-value"), but there is some evidence of intraspecific variation in some species but not in others. In mammals, there has not been a lot done on this, but there is some evidence in rodents and bats for geographical differences, albeit usually minor. And there is a 3% difference or so just between males (XY) and females (XX).

If so, is there a link between genome size and body size that would implicate genome size as a factor in determining an adaptive cline of body size? (for instance, in a system where latitude correlates with body size, such as in lions or canids?)

Body size is not linked to genome size even across mammals as a group. There is only about a 5-fold range in genome size in mammals, but many orders of magnitude in body size. Cell number is what matters, whereas cell size is what is affected by DNA amount. Body size does correlate with genome size in some invertebrates, especially when growth is determinate. However, this does not preclude your hypothesis, because mammals in the north have smaller erythrocytes than those in the south, presumably because they need higher metabolisms for thermoregulation, and erythrocyte size correlates with genome size in mammals even though differentiated mammal erythrocytes are enucleated.

By the way, we're working on some of these issues in my lab as we speak.

T Ryan Gregory said...

I think a clarification of "adaptation" vs. "constraint" is useful.

For example, if you find a correlation between genome size and latitude, it could mean more than one thing:

1) Genome size is tailored by selection on the organism level for that environment,

or

2) Genome size is set by processes at a lower level (intragenomic selection, mutation pressure) and then species with an incompatible genome size are precluded from some environments.

I would consider the first an adaptation, the second a constraint.

T Ryan Gregory said...

(Before anybody nitpicks, yes, the correlation could 3) be spurious).

Anonymous said...

thought we were past 'junk DNA' - ignorance for lack of current knowledge doesn't mean it's junk. Nothing like perpetuating dogma.

Anonymous said...

Like many people I answered something around 50%, because the entire system of DNA replication and transcription is optimized for this huge amount of DNA that we have, and I'm not sure what would happen if the total emount of DNA (regardless of the information encoded in it) would decrease too much.

For example: Can mitosis happen if there is only a teeny tiny bit of DNA? If it can't, then that's a pretty huge phenotype right there.