Statistical and high-throughput models of enhancer function and evolution

点击翻译按钮获取中文摘要

PROJECT SUMMARY Enhancers are an important class of noncoding loci regulating gene expression and play important roles in modulating diverse phenotypes and disease states. However, our understanding of the role of enhancers in phenotypic evolution is limited and we lack a detailed understanding of the relationship between sequence change within and between species, epigenetic states and variation in enhancer function. Moreover, we have few statistical models that allow researchers to connect evolutionary changes in enhancer sequences within and between species to phenotypic variation, and we often cannot unambiguously determine the causes of observed changes in evolutionary rate of enhancers along lineages. Finally, most studies of enhancer evolution thus far have studied only small numbers of enhancers and genome-wide assays of enhancer variation and function are rare. Here we propose to develop statistical models linking phylogenetic patterns of enhancer evolution with phenotypic variation between species, and to leverage within-species variation across multiple species – “comparative population genomics” – to disentangle the sources of rate changes observed in enhancers across species. We will also functionally test diverse enhancers on a large-scale, using the developing fore- and hindlimb of volant and flightless birds as a model of development and gene expression. Specifically, in Aim 1 we will extend a recently developed Bayesian phylogenetic model for detecting rate changes in noncoding DNA, phyloAcc, to improve its biological realism by incorporating stochastic gene tree heterogeneity and the ability to associate sequence change with both binary and continuous traits. Building on a novel data set of comparative gene expression and chromatin states across multiple species and developmental stages, we will also develop methods to associate genome-wide variation in chromatin states between species with binary and continuous traits. In Aim 2 we will develop additional statistical models to leverage information from sequence variation within species to better understand the evolutionary forces contributing to rate variation in noncoding DNA observed between species. The models developed in Aims 1 and 2 will be refined and made available to the broader community in a user-friendly format for use on diverse systems and species. In Aim 3 we will functionally validate large numbers of candidate enhancers identified in Aims 1 and 2 as having evolved new functions or found in altered chromatin states in the developing fore- and hindlimb of volant and flightless birds. Using high-throughput assays in chicken, emu and other birds we will study the relationship between within- and between-species sequence variation of enhancers and their ability to drive gene expression. Together these aims will provide a number of tools that will benefit the community of researchers using comparative genomics to understand links between genotype and phenotype, and will extend the kinds of phenotypic traits and genomic signatures that will inform this emerging paradigm.