mrakins said:
Turns out this might not even be true. I haven't read the article, but a recent paper (like this week, I think--I'm usually several weeks behind on the literature) gave evidence for the idea that the transposons are part of how our nervous system develops. Also, most of the differences between us and other primates (and birds and fish and trees) lies in the "junk DNA". Just another reason why no self-respecting biologist thinks that junk DNA is junk.
I remember the paper going by, and can't dig it up at the moment. If you can find the citation, please post it. The reason it made a splash, though, was that it was the exception rather than the rule. In the cases I know of, transposons, repetitive sequences and retroviral insertions tend to be associated with pathology more than with normal function. The best example I can think offhand of is the polyglutamine expansion associated with Huntington's disease.
I'm a little surprised that people really believe every base pair in their genome is there for a reason. If you look at gene structure, there is the part of a gene that encodes a protein (with exceptions like microRNAs, ribosomal RNAs and so on), the parts that regulate the expression of that protein, and then there is stuff in between. The part that encodes the protein is very strongly conserved between species, because you usually die if the protein is mutated. Certain regions of the regulatory sequence of the gene are also identical or very similar across species, because they are important for determining the place and time the protein is produced. But there has to be a certain amount of space between the various parts, so that DNA-binding proteins have room to bind and do their work. In that case, the intervening sequence doesn't make a fig of difference, it just has to be within a certain length range. Those sequences show no conservation across species, supporting the idea that the sequence doesn't make much of a difference.
There are also these bits of DNA called introns, within the parts that code for proteins. In some cases, there are sequences in the introns that are needed for regulation of expression. In others, the introns can be removed and everything is fine.
But what most "self-respecting biologists" refer to as "junk" are the dispersed repetitive sequences that are scattered throughout our genome. They were originally referred to as junk because there was no obvious role for them. At this point, their function is largely mysterious. It would be arrogant to say that they do nothing simply because we don't know their function, but it seems equally arrogant (and simply untrue) to believe that every base in our enormous genome serves a purpose.