Replies

Now, certainly, you must realize that it probably never was "junk" as you have previously argued. That argument , that it was "junk" had a reason, and that reason was in support of a certain theory. That it is not "junk" is, therefore, in contradiction to that certain theory.(that is, if the theory is to be credibly falsifiable as it is constantly alleged)

When did I ever say it had to be junk? Here's the exchange you & I had yesterday:

[JP-19] After all these years I still don't know what the evolutionary argument for the existence of junk DNA was supposed to have been. [I then remind you of my hypothesis for why introns would be advantageous to the evolution of complexity]
[AC-20] Really, now? It seems to me the argument was that, necessarily, "junk" was a basic component of a "random" process. In fact, most of DNA should be a complete wasteland of unused and "decaying" parts.
[JP-23] But junk DNA would be analogous to an unused organ. Since there's some nonzero cost to keeping & reproducing an unused DNA sequence, any deletion of a truly junk sequence should be slightly favored over time. [Unless the cost is negligible]...

This was my speculation for why we would have so much noncoding DNA when simpler organisms (bacteria) have none. This was before you provided the article that actually presented a bit of the history of mainstream thought on junk DNA, which showed that people on all sides of the issue (regarding introns at least) were using evolution as their starting point. From what I could understand, it looked like each hypothesis implied things we should see if their hypothesis was true. IOW, they all looked quite falsifiable to me.

BTW, here's a study that would be interesting to see the full results for. I don't know what exactly they're calling "junk DNA" - whether they include introns or just LINEs & SINEs, etc. But it's yet again a falsification test for one hypothesis over another:

Proc R Soc Lond B Biol Sci. 1992 Aug 22;249(1325):119-24.
Variation across species in the size of the nuclear genome supports the junk-DNA explanation for the C-value paradox.
Pagel M, Johnstone RA.
Harvard University, Cambridge, Massachusetts 02138.
The amount of DNA in the nuclear genome (the DNA C-value) of eukaryotes varies at least 80,000-fold across species, and yet bears little or no relation to organismic complexity or to the number of protein-coding genes. This phenomenon is known as the C-value paradox. One explanation for the C-value paradox attributes the size of the nuclear genome to 'junk' (typically non-coding) genetic elements that accumulate until the costs to the organism of replicating excess DNA select against it. Across species, organisms that develop at a slower rate should tolerate more junk DNA. Alternatively, junk DNA may function as a nucleo-skeleton to maintain the volume of the nucleus at a size proportional to the volume of the cytoplasm in the cell. Across species, the DNA C-value is predicted to vary with the nuclear and cytoplasmic volumes of cells. Previous studies have not been able to distinguish between the skeletal-DNA and junk-DNA explanations for the C-value paradox. We report a study of DNA content in 24 salamander species which does. The size of the nuclear genome is correlated with developmental rate even after the effects of nuclear and cytoplasmic volume have been removed. However, genome size is not correlated with cytoplasmic volume after controlling for developmental rate. These results support the view that junk DNA accumulates in the nuclear genome until the costs of replicating it become too great, rather than that it functions as a nucleo-skeleton.

This study raises the same issue of the cost of carrying the noncoding DNA as I did. So they show why junk DNA could accumulate in the genome in spite of normal evolutionary pressures toward efficiency. Couple this with Ichneumon's cite of the study where they actually removed junk DNA from the mice and they couldn't tell them apart, and you have a compelling case for them truly being junk.

Now, the cost of introns must be higher than simple psuedogenes & LINEs & SINEs, since introns are processed whenever a gene is transcribed - there's the whole spliceosome that needs to be in place to help excise them from the RNAs. So the cost of having introns should be higher than for the other noncoding DNA, so it would be more likely that they'd get selected out of the genome. Since we have so many introns in spite of this, that would argue for introns having some positive role to play that the other noncoding regions wouldn't have.

So far all I see are lots of good, falsifiable hypotheses getting generated for why introns exist, all using the truth of evolution as a starting point. What's not to like? And where are all the falsifiable creationism-inspired hypotheses & studies showing why introns exist?

The following doesn't have anything to do with the debate per se, but in the interest of completeness:

This study raises the same issue of the cost of carrying the noncoding DNA as I did. So they show why junk DNA could accumulate in the genome in spite of normal evolutionary pressures toward efficiency. Couple this with Ichneumon's cite of the study where they actually removed junk DNA from the mice and they couldn't tell them apart, and you have a compelling case for them truly being junk.

I should clarify something: It looks like the junk DNA they removed from the mice really is junk in the short term. I still think there's a circumstantial case for junk DNA being useful on macroevolutionary timescales, by providing the buffer between functional subunits that ensures their integrity during partial gene duplications. The test for this would entail some kind of statistical analysis of exons & introns vs. what would be expected if rearranging protein parts along functional-unit boundaries was not selected for.

Another thought I had: For each new processed pseudogene or some other stretch of junk that an individual organism acquires, would the extra cost of the new junk really be felt by selection? Each new piece of junk probably doesn't rise above the "selective noise level" - the disadvantage it gives to the organism vs. their competing cohorts would be negligible. So within a species, each new piece of junk would just be another neutral mutation & could easily get fixated in the gene pool.

It wouldn't be until the species had to compete with another related species that a smaller junk DNA burden might be selected for (because there would be more time for many different insertion events to get fixated at different rates in the two competitors). Eh, even still the difference might be small.