Transposons.

I've never heard of them being useful to an individual organism. They are generally considered to be
selfish genes. Come to think of it, some are known to carry antibiotic resistance genes...somehow
promoting the horizontal spread of these genes among bacteria. Sorry, but I'm not up on transposons.

However, I have heard proposals that they can be helpful to a linage. The idea is that they can make
mutations that may be beneficial on an evolutionary timescale. For example, they can jump into a
gene and knock it out (or modify its expression), but then excise themselves from that gene at a
later time, allowing it to regain (near) regular activity... I can't say exactly how that would be
beneficial, but it is neat. Similarly, transposases might be involved in some large-scale genome
rearrangements. Again, I can't say exactly how that would be beneficial, but it surely has
evolutionary consequences. Anyway, I think this whole line of reasoning regarding the adaptiveness
of transposons is largely speculative.

They are hella good for researchers though!

-adam Malcolm said:

About 30% of the human genome is considered to be the remains of transposable elements
(transposons), jumping genes which insert copies of themselves around the genome.

Is this DNA just detritus / parasitic, or are there any examples of it being of some use to
the host?

Well, there is indeed a big controversy on whether reapeated elements (not
just transposons) carry with them some selective advantage or if they are just
neutral or slightly deletereous. The most widely held view is that repeated
elements can become deleterious when they accumulate in great quantities,
hence, imposing an upper limit to the genome size of an organism. Proponents of
the selfish gene model for repetitive elements state that these DNA sequences
share special properties that lead them to a rate of replication within the
genome that is higher than for classical genes (when normal genes duplicate
within a genome they can form pseudogenes or orthologs, etc.), and they
propose a model where a selfish gene would have certain 'growth rate' that
will determine how fast it propagates, and there would be some limit imposed by
selection, which can be think of a 'K' value in a logistic model of population
growth (well, this is just an analogy, not an strict mathematical comparison).
Proponents of the selfish gene model also state that some repeated elements
might adquire selective advantanges that increase the organism fitness, but
this is not a characteristic of all selfish genes, for example, the way
inmunoglobulin gene are shuffled and generated resemble selfish genes
mechanisms. On the other hand, there are the functionalists that state that
ALL repeated DNA has some function, for example in chromatin structure in the
chromosomes, or as importat regulatory elements in gene expession. I consider
the selfish proposition most atractive, specially because it has some elegant
mathematical models. For a recent, and more comprenhensive review on this
subject you can check:

Petrov (2001) Trends in Genetics, 17:23-28

Malcolm said:

About 30% of the human genome is considered to be the remains of transposable elements
(transposons), jumping genes which insert copies of themselves around the genome.

Is this DNA just detritus / parasitic, or are there any examples of it being of some use to
the host?