There is a preprint in the website of The American Journal of Human Genetics titled "Genetic variation in the CCL18 - CCL3 - CCL4 chemokine gene cluster influences HIV-1 transmission and AIDS disease progression." The title is a mouthful, but the short of it is what we've known for a long time, that human genetic variatian responds differently to HIV infection (or the risk of infection). This is surely going to be important, not because the science is a priori killer, but because AIDS is a big public policy issue. Back in the 1990s some people were talking about HIV resulting in the extinction of the human race (OK, I'm mostly talking about my teachers), but real knowledge of evolution would have implied that this is ridiculous. Plagues and pandemics come and go, but the species always bounces back, and we are a numerous large mammal so operational immunity is almost certainly going to be found amongst a small minority (as it has been). In any case, since the paper is a preprint there isn't an abstract. Rather, the major points seem to be this: some SNPs are correlated with faster disease progression, and Europeans are also characterized by lower polymorphsim and greater linkage disequilibrium than African Americans on the region of the genome of interest. The linkage disequilibrium manifests in "haplotype blocks" that may be the product of recent powerful selection on loci around which recombination has not had time to break apart the associations generated by the hitch-hiking effect.