To ring out 2009, a brief mention of some studies that reinforce the conclusions mentioned in this essay. The basic theme – it looks more and more as if poly(A) site choice can vary for many, many genes in humans, and that a general trend towards shorter 3’-untranslated regions(due to selection of poly(A) sites nearer the 5’ end of the mRNA) is seen in rapidly growing cells (such as stem cells and cancer cells). The consequences of this seem to be significant, as it seems as if shorter RNAs are more abundant, owing to an absence of RNA sequences that destabilize the mRNA. In other words: shorter 3′ UTRs = higher expression = more growth/less regulation.
The means by which such global trends in poly(A) site choice may be accomplished are not clear; this will undoubtedly be one of the areas of interest in the field of polyadenylation over the next few years.
Posted by Arthur Hunt 
The RNA Underworld 