The paper on the ability of humans to track smells contained a couple interesting references on the human specific loss of some olfactory genes (see an old post here for an interesting correlation between the loss of these genes and the rise of color vision). Apparently, some genes have both functional and non-functional versions that still segregate within the population. So do different populations have receptors for different scents?
The last link is to a paper that typed 51 olfactory receptor genes in 189 people. It’s a little unclear where these 189 people come from (though they do refer to African-Americans, so it’s apparently a sample of Americans); it would be interesting to type them in a large number of populations (the human diversity panel, perhaps?) and give that hypothesis a whirl.
However, see this paper for an argument that humans actually have a pretty good sense of smell, and some possible explanations for why the loss of olfactory receptors doesn’t mean the loss of the sense:
[M]uch of the olfactory system can be removed with no effect on smell perception. The olfactory receptor genes map topographically onto the first relay station, a sheet of modules called glomeruli in the olfactory bulb. Up to 80% of the glomerular layer in the rat can be removed without significant effect on olfactory detection and discrimination. If the remaining 20% of the glomeruli-and the olfactory receptor genes they represent-can subserve the functions of 1,100 genes, it implies that 350 genes in the human are more than enough to smell as well as a mouse.