junk DNA

Ian A. York iayork at panix.com
Mon Apr 17 18:10:37 EST 2000

Let me take another run at that last post.  You hit several of my pet
peeves all at once, and I took it out on you instead of the proper target.

There are several problems with the Discover article.  Mainly they are the
result of Discover being a journalist-run magazine, rather than a
scientifif magazine.  The attitude is very different; journalists have it
drilled into their heads that "balance" is important, and most seem to
believe that, for example, "balance" means giving equal time to tiny
minority camps.  That's the mindset that leads to giving half an article
to antievolutionists.

The other problem with the approach is that it fails to give the proper
background, preferring to make the focus of the article seem more
important by ignoring the full context.  In this case, the article
discussed a five-year-old publication, ignoring the fact that no one has
followed it up or confirmed it; and implied that "the" function of alu
elements has for the first time been explained.

The key context the article ignored is that many, many functions for alu
elements have been proposed, many with at least as much evidence as the
one put forward there.  That is, this is one of many proposed functions
for alu elements, not a breakthough explanation.  There are many good
reviews on them; you can start with FEBS Lett 1997 Nov 3;417(1):1-5 (Alu
sequences.  Mighell AJ, Markham AF, Robinson PA.)

And that raises a second problem.  Although the selfish DNA hypothesis for
junk DNA is not invariably accepted, and certainly isn't invariably
accepted as a general explanation for all junk DNA, almost everyone
accepts that transposons, and specifically alu elements, *are* essentially
selfish DNA.  It's clear that they replicated themselves explosively at a
relatively recent time in our evolution (and are continuing to expand even
today).  No one, that I know of, believes that this expansion of alu
elements (and other transposons) was a selected process.  When people
propose functions for alu elements, they are proposing *co-opted*
functions--not primary functions that led to their selection.  That is,
although alu elements are generally accepted to have expanded for no
reason, some of the individual elements have fortuitously affected other
"real" genes in various ways.  In other words, alu elements have functions
*in spite of being* junk DNA originally; they do not have functions that
prove they are *not* junk DNA originally.

By the same token, and strongly arguing against any positive selection for
alu elements, they are frequently implicated in negative events.  It has
been suggested that alu elements are responsible for dozens of human
genetic diseases.  Again, these are fortuitous events, that happen to have
negative impacts, as on other occasions the random effects of alu elements
turn out to be positive.

Note that affecting evolution is not a function, because that's not how
evolution works; that implication depends on the belief that evolution is
a directed function--that there is selection for a mutator phenotype, to
put it technically.  A great deal of theoretical and experimental work has
not completely ruled out any possibility of a mutator phenotype being
selected for, but it does make it clear that it is an extremely rare
event, and very weakly selected.  Therefore any effect on primate
evolution is at most a co-opted effect, not a true selected function.

Finally--well, not finally, but it's all I'm going to say on this
subject--the article conflates alu elements and junk DNA in general.  Alu
elements are a minority of total transposon elements, which are in turn a
minority of "junk" DNA.  They are not representative of junk DNA in
general, and making a statement about alu elements doesn't allow one to
make a statement about junk DNA.

    Ian York   (iayork at panix.com)  <http://www.panix.com/~iayork/>
    "-but as he was a York, I am rather inclined to suppose him a
     very respectable Man." -Jane Austen, The History of England

More information about the Cellbiol mailing list