Alright, I'm going to write about an interesting review article on gene duplication that I just read, on the grounds that I don't have anything better to write about, so I might as well node blog my homework.

So, traditionally evolution is considered to progress by small mutations accumulating over time, with beneficial mutations being subject to positive selection and harmful mutations being subject to negative selection. A problem arises, however, if we consider how it is that a new function is supposed to arise in a gene by such stepwise mutations: any mutation that confers some new function is likely to abolish the old function. Hence, it would seem that new functions could not arise from genes responsible for the production of essential proteins, since the arising of these new functions would abolish the old, essential functions, and therefore would be selected against.

Enter gene duplication. This is a process where some error in recombination (or some other process, such as the function of a retrotransposon) results in the duplication of a gene, some section of a chromosome, an entire chromosome, or an entire genome becoming duplicated. This resolves the above problem because there are now two copies of the gene present, one of which can retain the original function while the other can accumulate mutations without harming the organism.

Now, mutations are much more likely to abolish the function of the gene than to result in new function. Therefore, in the majority of gene duplication events, we would expect that one copy would be preserved while the other copy becomes nonfunctional due to the accumulation of deleterious mutations, thereby becoming a pseudogene. In some rare cases, we would expect that one copy would be preserved while the other copy acquires a mutation that confers some new, beneficial function, leading to both copies of the gene being subjected to positive selection. Wikipedia says as much:

This is because with two copies of a gene present, mutations in just one copy of the gene often have no deleterious effect on the organism; thus, the second copy is free to "explore" the sequence space by mutating randomly. The duplicate gene may either (a) acquire mutations that lead to a gene with a novel function or (b) acquire deleterious mutations and become a pseudogene.

However, when we examine the prevalence of preserved duplicate genes in extant genomes, we find that duplicate genes are preserved at a rate much higher than would be predicted based upon the assumption that most duplicated genes will become pseudogenes. This is because it is mistaken to consider genes as being composed of only a single unit, and, as such, there is actually a third possibility for the fate of a duplicated gene which the traditional model (and Wikipedia) omit.

The article Preservation of Duplicate Genes by Complementary, Degenerative Mutations by Force et al. [Genetics. 1999 Apr; 151(4): 1531-45] demonstrates this third possibility in this figure:

The gene in this figure has a large coding region and 4 small regulatory regions that control expression in different tissues. The two possible fates discussed above are shown in the two leftmost columns in the figure: nonfunctionalization, or the loss of function of one copy whereby it becomes a pseudogene, and neofunctionalization, or the gain of new function of one copy. The third possibility is subfunctionalization, whereby the loss of some discrete subfunction in one copy of the gene is accompanied by the loss of some different discrete subfunction in the other copy. By this mechanism, both copies become necessary for retention of the original function. This process could explain why we observe preservation of both copies at a higher rate than would be predicted when taking only nonfunctionalization and neofunctionalization into consideration.

Another interesting thing about this hypothesis is that, in the example above, different versions of the duplicated gene are expressed in different tissues. This could allow for changes to occur which are specific to those tissues; therefore, this process could be a mechanism whereby developmental changes might evolve.

Anway... it's an interesting paper.

Vote Greg Chaimov for City Council.

It seems that LiveJournal, who once upon a time included in their Social Contract the promise "to never offer advertising space in our service or on our pages" has decided to start offering "sponsored communities" and "sponsored features" with advertisements that will be displayed to all users — even paid and permanent users. It would seem that Six Apart holds non-orthodox definitions of the words promise and never.

Everyone needs to go tell them how fantastic their plan isn't.

You don't say...