Evolutionary genetics of cis-regulatory variation in humans.

摘要

Functional genetic variation is the raw material for evolutionary change, but the molecular nature of functional variation, and the evolutionary mechanisms that generate and maintain it, are poorly understood. Almost nothing is known about cis-regulatory variation in nature, despite its status as one of the major classes of functional variation. In this thesis, I focus on cis-regulatory variation in an important model system for studies of functional genomics, humans. I synthesize available data on the molecular nature and population genetic characteristics of human cis-regulatory polymorphism, and I investigate the role of evolutionary forces---mutation, natural selection, and genetic drift---in shaping the observed pattern of variation.; Case studies of three exceptionally well characterized functional cis-regulatory variants allow for a dissection of the evolutionary history of regulatory mutations. At all three loci, IL4, MMP3, and PDYN, we find strong evidence for positive selection increasing the frequency of derived mutations. In the case of IL4 , a new transcription factor binding site has been favored in some parts of the world, while the alternative allele has been favored elsewhere. At MMP3, a mutational hotspot resulted in the loss of a binding site for a transcriptional repressor, and ancient natural selection favoring the loss has altered the expected distribution of phenotypes within contemporary populations. PDYN exhibits a remarkable pattern of strong positive selection fixing cis-regulatory mutations during the descent of modern humans from our last common ancestor with chimpanzees, and an additional layer of positive selection acting heterogeneously among human populations. An investigation of the rates of substitution across a large dataset of functional human binding site polymorphisms suggests that rapid evolution may be a fundamental characteristic of the sites at which cis-regulatory polymorphisms occur in humans. A role for hypermutability and hyperselectability in attracting functional variation has implications for the direction of medical genetics and for an issue of central interest to evolutionary biologists---what are the causes of phenotypic evolution?