Most human proteins (functional units of biological machineries in a human cell) contain evolutionarily conserved protein domains (components of a functional unit). The sequencing of the human genome has resulted in a phenomenal increase in the identification of such protein domains, but their molecular functions or their roles in human cellular physiology are poorly understood. Knowledge of the three-dimensional structure or shape of a protein domain helps infer its molecular function in biology. This project aims to develop a new structure-based paradigm for genome-wide profiling of molecular functions of protein domains. To achieve this goal, the researchers will use a family of protein domains, called the bromodomain, as a model system. The bromodomain functions to interact with acetylated lysine (a chemically modified natural amino acid) on a protein. Such a molecular interaction plays a fundamental role in establishing a network of dynamic protein-protein interactions in a wide range of cellular processes including gene expression. Despite their fundamental functions in human cell biology, target selectivity of human bromodomains, a large family containing more than 170 members, is not understood. Notably, Saccharomyces cerevisiae (brewers' yeast) has only 15 bromodomains that functionally represent different subsets of the much larger human family. As such, new knowledge of target selectivity of the yeast bromodomains can be extended to understand the functions of the human bromodomains. Towards this goal, this project employs a new strategy of combining structure-based experimental and computational methods to profile target selectivity of bromodomains from yeast to human.
The outcome from this research is expected to yield mechanistic understanding of the fundamental and complex cellular functions of biological machineries in a wide array of cellular processes. Successful establishment of new computation-based prediction methodology using combined and iterative experimental and computational approaches being developed in this project will greatly enhance our ability to use the highly valuable information coded within genome sequences. In addition, this project will be used as a vehicle to provide interdisciplinary graduate training, both in the researchers' labs and through courses offered through the Biophysics, Structural Biology and Biomathematics Multidisciplinary Training Area in the Graduate School of Biological Sciences.
