Cis-regulatory elements and human evolution

Abstract

Modification of gene regulation has long been considered an important force in human evolution, particularly through changes to cis-regulatory elements (CREs) that function in transcriptional regulation. For decades, however, the study of cis-regulatory evolution was severely limited by the available data. New data sets describing the locations of CREs and genetic variation within and between species have now made it possible to study CRE evolution much more directly on a genome-wide scale. Here, we review recent research on the evolution of CREs in humans based on large-scale genomic data sets. We consider inferences based on primate divergence, human polymorphism, and combinations of divergence and polymorphism. We then consider 'new frontiers' in this field stemming from recent research on transcriptional regulation.

Figure 1

(A) Frequency as a function of time for hypothetical mutations experiencing neutral drift (gray), weak negative (green), strong negative (blue), or positive (orange) selection. The plot assumes a new mutation occurs in a single individual in the population at time 0. Neutral drift typically causes mutations to be lost (lower gray fork) but occasionally drives them to fixation (upper gray fork). Negative selection essentially guarantees eventual loss, but if it is sufficiently weak (green plot), mutations may segregate at low frequencies for some time. Positive selection (orange plot) causes mutations to reach fixation at higher rates than neutral drift. Notice that the time until fixation or loss is substantially reduced for mutations under strong selection (positive or negative), implying that they are unlikely to be observed in a polymorphic state. (B) Steady-state numbers of invariant sites, low frequency (derived allele) polymorphisms, high frequency polymorphisms, and fixed differences under neutral drift, expressed as hypothetical percentages of nucleotide sites. These represent equilibrium frequencies for the process depicted in panel (A) for a given divergence time, assuming a steady influx of new mutations. Positive selection (orange arrows) increases fixed differences, reduces invariant sites, and reduces polymorphisms. Strong negative selection (blue arrows) reduces fixed differences and polymorphisms and increases invariant sites. Weak negative selection (green arrows) is similar but allows some low frequency polymorphisms to remain. (C) Phylogenies with branch lengths proportional to rates at which fixed differences occur along lineages. Positive or negative selection can be identified by significant increases or decreases, respectively, in the fixation rates relative to the expectation under neutral drift. Different likelihood ratio tests can identify lineage-specific or recurrent/homogeneous selective pressures. (D) Scatter plot of polymorphism vs. divergence rates under neutral drift, generated by simulations based on parameters reflecting real human populations [45] (black points). Points represent different loci with variable mutation rates. Colored points show hypothetical positions of loci under positive (orange), strong negative (blue), and weak negative (green) selection. Notice that positive and negative selection are distinguishable by their joint effects on polymorphism and divergence rates, but not by polymorphism rates alone. (E) 2 × 2 contingency table used for McDonald-Kreitman (MK) test for selection on a cis-regulatory element (CRE). The test evaluates the probability of the observed data under the null hypothesis that the relative polymorphism and divergence counts are independent of the labels “reference” (sites putatively under neutral drift) and “CRE”. The classes of sites are chosen to be similar to one another in other respects to avoid potential biases from mutation rate variation and demography. Rejection of the null hypothesis therefore implies a departure from the neutral expectation of equal fixation rates. Note the connections with the visual representations used in panels (B) and (D). The MK test can be thought of as comparing the relative heights in panel (B) of the first bar and the next two bars combined, for reference vs. CRE sites (see arrows). It can also be thought of as testing for extreme departures from a diagonal line in panel (D) running through the neutral points from bottom left to top right. In this example, the counts reflect an excess of fixed differences in the CRE, suggesting positive selection. Notice that strong negative selection is not a problem for the MK test, because it reduces the effective mutation rate, but weak negative selection can bias the test by partially canceling the effects of positive selection.

Figure 2

Some of the many factors that may influence the evolution of cis-regulatory elements.