EAGER: Towards Understanding the Information-Theoretic Nature of the Human Epigenome

Johns Hopkins University, Baltimore MD

Investigators

Abstract

Cells use an elaborate system to determine when and where specific genes will be expressed during development and differentiation as well as in response to environmental conditions and stimuli. This system is overlaid on the DNA in the form of heritable epigenetic marks that control gene expression by modifying the structural organization of the genome without changing the DNA sequence. Epigenetic marks can silence genes or activate them by adapting chromatin (a DNA/protein complex which forms chromosomes within the nucleus of eukaryotic cells) to distinct states that repress or stimulate gene activity and this can drive genetically identical cells to behave differently from each other. There is now ample evidence that aberrant epigenetic regulation can lead to disease. However, and in sharp contrast to gene mutations, epigenetic alterations can be reversed. It is therefore believed that research in understanding the human epigenome can lead to novel and highly effective therapeutic strategies for many human diseases, such as cancer, diabetes, and Alzheimer?s. A formal pursuit of the proposed research will provide a solid foundation for developing fundamentally different methods for the modeling, quantification, and analysis of epigenetic information, as compared to rather crude mathematical and computational methods currently used in the literature. This could potentially have a major impact on the area of epigenetic science, as well as on medicine and society at large, which could lead to new biological discoveries towards understanding the role of epigenetics in development, disease and aging. The main goal of this research is to develop a novel approach for understanding the informational structure and properties of the human epigenome by using well-grounded biological assumptions and principles of statistical physics and information theory. The investigators will develop methods for quantifying epigenetic stochasticity, as well as discern and analyze epigenetic discordance between biological samples, providing new and exciting ways for studying the role of epigenetic regulation in disease and aging. By viewing the process of transmitting epigenetic information during cell division as a communication system, the concept of a methylation channel is introduced, which can be characterized by its capacity, dissipated energy, and input/output entropy. Preliminary results using real epigenetic data have demonstrated an intriguing connection between chromatin organization and informational properties of methylation channels. This shows that a merger of epigenetic biology, statistical physics and information theory may lead to fundamental insights into the relationship between the informational properties of the epigenome and nuclear organization in normal development and disease. Successful completion of the proposed work will transform the way epigenetic information is modeled, quantified, and analyzed, and lead to powerful methodologies for understanding the information theoretic content of the human epigenome and its role in development, disease, and aging.