Chromosomal organization and chromatin structure play a major role in the regulation of gene expression, with important functions in development, differentiation, and cancer. Our long-term goal is to understand their genome-wide effect on gene regulation by developing and applying probabilistic methods for identifying regulated chromosomal domains (CDs) - physical clusters of co-regulated genes, and the mechanisms that control them, from genome-wide expression profiles and other genomics data. Such a genome-wide study of CDs can provide the missing link between the emerging field of chromatin structure and regulation and the transcriptional readout that this struct ure is thought to direct. As chromatin and its modifications are implicated and targeted in a rang e of human cancers, this link to their transcriptiona I effect may have important diagnostic and thera peutic implications.
Our specific aims are to: 1. Create a comprehensive map of chromosomal domains. We will develop probabilistic methods for identifying CDs from gene expression profiles and other genomics data, and an accompanying statistical framework for characterizing the function and behavior of CDs across different biological conditions. 2. Identify candidate molecular mechanisms that regulate CDs. We will construct detailed mechanistic models of CD regulation that integrate heterogeneous types of genomic data, including gene expression, DMA sequence, and histone modification data. These will suggest specific testable hypotheses regarding the sequence elements, regulatory proteins, and epigenetic features that are involved in CD regulation. 3. Elucidate the evolutionary history of chromosomal domains. If CDs serve an important function, we expect part of their organization to be conserved across organisms. We will develop a comparative genomic framework to identify evolutionary conserved CDs and sequence elements that are conserved within or at CD boundaries, and use these results to characterize the role of CDs in genome evolution. 4. Understand the role of chromosomal domains in cancer. Chromosomal aberrations are a hallmark of cancer cells. We will develop a probabilistic model of chromosome aberrations that integrates expression and comparative genomic hybridization (CGH) data to distinguish between mis-regulation of CDs and gross chromosomal changes, suggesting a unified model for the role of regional regulation in cancer.
Showing the most recent 10 out of 40 publications
