There’s been a lot of recent interest in sequences that are highly conserved between humans and other mammals (and even other non-mammalian vertebrates). These sequences are thought to be under purifying selection, which prevents the accumulation of substitutions after two evolutionary lineages diverge. We cannot rule out, however, that the sequences are conserved by either pure chance (ie, in a large enough dataset, there will be a substantial amount of outliers that are still part of the same distribution) or because they have a lower mutation rate.
David Haussler is one of the leaders in the study of ultraconserved sequences in the human genome. In an attempt to understand the evolutionary dynamics of the ultraconserved sequences, Haussler and colleagues studied the patterns of polymorphism in these regions. Population genetics theory predicts that, if natural selection were responsible for the conservation of these sequences, they should also have low levels of polymorphism because natural selection would be purging newly arising mutations before they reach appreciable frequencies.
Haussler and colleagues found that there is a marked excess of rare alleles segregating in human populations at these ultraconserved elements. New mutations in ultraconserved elements stick around long enough to be found in a couple of individuals, but they are removed from the populations. In fact, this excess of rare alleles is even greater than that in protein coding regions, indicating that the ultraconserved elements are under more purifying selection than regular genes.
Katzman S, Kern AD, Bejerano G, Fewell G, Fulton L, Wilson RK, Salama SR, and Haussler D. 2007. Human genome ultraconserved elements are ultraselected. Science. 317: 915. doi:10.1126/science.1142430