Simon Gravel has a new preprint up on bioRxiv, When is selection effective? It’s a preprint, so has to be thought of as a work-in-progress. From my perspective it’s interesting because it combines analytic methods along with simulation in an attempt to sharpen intuitions about the power of selection to modulate genetic load. Issues relating to load matters because there have been empirical results and arguments about the differences between human populations due to findings from genomics over the past 10 years (e.g., Europeans have higher load than Africans because of lower long term effective population size). More generally I believe that the interplay of selection and drift across natural history are relevant for conservative genetics.
These results seem to imply that using realistic models of human demographic over the past ~100,000 the differences in load should be relatively minor. Interestingly the power of selection on recessive alleles of large deleterious effect actually becomes stronger in bottlenecked populations, presumably because of exposure of homozygotes. This is obvious in hindsight. In contrast weakly deleterious alleles are more efficiently purged in the larger effective population size of Africans.
The main thing I took away from the preprint is the emphasis on the long term population history and its impact on genetic load in a given generation. It strike me that this is why simulation methods are so persuasive, as the combined effects are indeed subtle.
Citation: When is selection effective?, Simon Gravel, http://dx.doi.org/10.1101/010934