This research combines computer simulation (energy optimization/expansion techniques, Monte-Carlo calculations, and normal-mode analysis) with a new naturally discrete model of DNA to examine the configurations and properties of supercoiled DNA, a topologically constrained form of the double helix subject to higher-order folding and compensatory strand twisting and the way that DNA exists in living cells. Base sequence-dependent features of the long, threadlike polymer are incorporated in the theory and treated by numerical simulations. By combining analytical studies with computer simulations, we obtain complementary information and have a series of built-in checks and balances for assessing the significance of our findings. The computational results stimulate new theoretical developments, which in turn can be used to assess the validity of the calculations. The effects of the sugar-phosphate backbone and local chemical environment are treated implicitly with knowledge-based potentials extracted from high-resolution structures of double-helical DNA and, in representative cases, with explicit treatment of long-range forces. Our immediate goals are to study the sequence-dependent biophysical properties of DNA minicircles and loops and to establish the physico-chemical basis of in-vivo looping. The proposed studies aim to clarify the role of primary chemical features (base sequence), ligand binding (proteins), and levels of supercoiling (imposed constraints on base-pair positioning) on the overall folding of the double helix. We will extend our studies of DNCA loops tethered to the Lac repressor protein with the aim of gathering new insights into the role of the molecular assembly in the regulation of transcription and its potential applicability as a model of eukaryotic insulators. A second goal is to gain insight into the effect of DNA sequence and supercoiling on protein binding and consequent cellular function. Other issues to be addressed include: (1) the role of local sequence-dependent structure and ligand binding on large-scale configurational transitions of spatially constrained duplexes; (2) the competing effects of multiple proteins on the configurational properties of supercoiled molecules; and (3) the interplay of local and global structure in supercoiling dynamics.

This research examines both atomic-level and mesoscopic-level protein:DNA interaction, to gain insight into molecular basis of protein-DNA binding specificty and affinity. Our approach endeavors to address the sequence-recognition problem from the perspective of DNA and attempts to understand the interactions between DNA and protein molecules in the context of DNA helical conformation. We also developed a relational database which enables us to search for contact patterns and associated nucleic-acid and protein structures using a broad range of chemical and structural criteria. It also allows us to study the conformational context and structural consequences of these contacts, which are quantitatively measured by various DNA structural parameters. Our preliminary findings show that shear deformations (specifically base-pair Slide) play a much more important role in the formation of nucleosomal DNA than heretofore thought. Moreover, large Slide deformations imposed on DNA by the histone proteins at the sites of minor-groove kinks appear to facilitate nucleosome positioning. Indeed, the cost of the “kink-and-slide” conformation found at sites of close histone-DNA contacts varies significantly among the 10 unique dimers (with the lowest costs at TA and CA:TG steps), suggesting a new mechanism of sequence-directed nucleosome positioning. Our observations highlight the importance of accounting for the contributions from each type of dimeric deformation (and the correlations of conformational variables) in predictions of nucleosome positioning. We also discovered apparent sequence-dependent build-ups of chemical species on the surface of both the base pairs and the sugar-phosphate backbone of DNA in diversified protein families.

Diversity of RNA structures stems from a large number of non-canonical base pairs formed by base-base and base-backbone interactions. In order to understand the occurrence and distribution of Watson-Crick and non-canonical base pairs in RNA, we have developed a database of RNA Base-pair Structures (BPS) to provide efective tools for base-pair search and analysis. High-resolution RNA crystal structures are analyzed with the 3DNA software package to obtain information about base pairs, higher-order base interactions, and helical organization. A parser package X3DNAParser has been created to automate structural analysis and data retrieval. The BPS database can be searched via a web interface that makes use of dierent types of search patterns. A systematic survey of currently determined high-resolution RNA structures identifies 28 predominant base-pairing patterns involving AA, AC, GA, AU, GC, GG, GU, UU interactions in terms of distinct hydrogen-bonding patterns and base-pair orientations. The analysis of isosteric base pairs that are geometrically similar provides information concerning sequence covariation. The 28 dominant base pairs have been used as "seed' patterns to identify isosteres. Isosteres in each clusters play a similar role in three-dimensional structural contexts.

From Yurong Xin's PhD Thesis.

The space of conformations of RNA nucleotide steps in the ribosome has been analyzed by Schneider et al using its backbone torsions angles and classifying them using techniques commonly used in X-ray crystallography which have been grouped into ARNA-type and non ARNA-type dinucleotide steps using a simple lexicographical clustering method. In the non-ARNA group they found 18 subgroups which are our starting point for implementing novel techniques used to define the topological boundaries of the periodic table of the elements from multidimensional elemental property vectors. The techniques used to obtain elemental periodic tables from their properties can also be used to produce molecular periodic tables which have and underlying rigourous topological classification, these elemental units of molecular structure can be related to what is know as RNA motifs, or perhaps even further to RNA folds.

From Mauricio Esguerra's PhD Proposal.