Wednesday, November 22, 2006

The HapMap and copy number variation   posted by p-ter @ 11/22/2006 03:05:00 PM

In the HapMap database are genotypes for over 3 million single nucleotide polymorphisms (SNPs) in 270 people-- 90 people of Western European descent, 90 people from Nigeria, 45 from Japan, and 45 from China. This is a spectacular resource for all sorts of population genetic, medical, and evolutionary studies. Yet SNPs aren't the only way people vary-- duplications and deletions of genome material are also important (as are other changes). Now, for those same 270 people who make up the HapMap, a catalogue of the duplications and deletions (or really, all variation in copy number) segregating in the populations has been generated. This is reported in the most recent issue of Nature.

Of note, they find that the actual quantity of DNA affected by these copy number variants covers a full 12% (!) of the genome. That is a huge amount of variation (we're all genetically 99% the same, right?), but is it functional? The authors note that copy number changes tend to stay away from known genes, but there are a number of examples of functional copy number changes, including ones that influence succeptibility to the HIV virus or glomulonephritis (whatever that is). I expect these sorts of variants to be very important in complex traits-- they provide the raw material for subtle changes in regulatory networks, which will then affect subtle phenotypic changes.

Another interesting story is that of the MAPT locus. This locus is marked by an inversion on one haplotype that has been under selection in Europeans. The two haplotypes at the locus are extremely diverged, which to some suggest another introgression event from Neandertals. An inversion that remains polymorphic in a population is interesting, because individuals heterozygous for the inversion are expected to have reduced fertility (recombination in the area of the inversion leads to imbalanced chromosomes). The apparent positive selection on this locus is, in that context, a bit puzzling. I'm not sure the authors realize this, but they might have solved this problem--if the duplication is the cause of the increased fitness, it would conteract the deleterious effects of the inversion. Unfortunately, there's not a lot of detail here, and it's not even clear that the duplication is on the positively selected haplotype.

Finally, there seem be some copy number variants that are have very different frequencies in the populations examined, suggesting possible differential selection. Among them is the variant known to decrease succeptibility to HIV and another known to play a role in androgen metabolism.

So the HapMap cell lines pay off once again and the human genetics community reaps the rewards; it's hard to believe that a number of people were against the construction of this resource.