A short paper late last year by Danielle Jones and John Wakeley examines the influence of gene conversion on the linkage disequilibrium that results from positive selection. It ends:
Then, among 100 chromosomes that all possess the selected allele, we would expect to see about four of these aberrant haplotypes, and the chance that all 100 chromosomes would show the classic, recombination-only sweep pattern would be 0.9625100 ≈ 0.024. Thus, it is possible that many selected loci have been missed in the recent genomic scans for selection.
Just thought I'd point out this example, although its importance is probably not too great. Low-frequency selected alleles are probably a much more important reason why current scans lack statistical power to find recent selection. There should be a lot more than have currently been identified in any scan, since the samples only number in the hundreds.
References:
Jones DA, Wakeley J. 2008. The influence of gene conversion on linkage disequilibrium around a selective sweep. Genetics 180:1251-1259. doi:10.1534/genetics.108.092270
