DescriptionThe mechanical properties of DNA play a key role in its biological processing, determining how the long, thin, double-helical molecule responds to the binding of proteins and functions in confined spaces within a cell. In eukaryotes, about 75-90% of genomic DNA exists in the form of nucleosomes, which are the fundamental units of DNA packaging in chromatin and the primary determinate of DNA accessibility. The structure of chromatin undergoes various changes that depend, at least in part, upon the requirements of gene expression and other functional environments. The dynamics of DNA packaging in chromatin is thus fundamental to numerous biological processes.
The flexibility of DNA is important in packaging DNA over lengths comparable to its persistence length during genetic processing and the sequence-dependent properties of DNA determine the positioning of nucleosomes in the genome and the sites of binding of enzymes and transcription factors. In addition, understanding the correlation between DNA flexibility and histone-DNA interactions inside the nucleosome is essential for unraveling currently unsolved mechanisms of gene regulation. Furthermore, although many experimental techniques have emerged to examine the overall structure of chromatin fibers, the internal arrangement of DNA and histones remains unclear. Thus an appropriate computational model able to incorporate experimental observations is key to interpretation of the folding and unfolding of chromatin.
The major goal of this thesis is to understand some of biophysical mechanisms involved in the packaging of DNA into chromatin using computational techniques at multi-scales: (i) to determine the sequence-dependent flexibility of DNA by developing DNA deformation analysis tools and databases; (ii) to design DNA spatial configurations using knowledge-based Monte-Carlo sampling; (iii) to map protein-DNA recognition inside nucleosomes in terms of realistic molecular treatments; and (iv) to interpret the internal structure of chromatin fibers and examine chromatin looping using novel modeling and simulation methods.