| The structure-function relationship is an important theory in nucleic acid chemistry, as the three-dimensional structure of nucleic acids can often help elucidate its function. In order to compensate for the disparity between known primary and tertiary structures, accurate secondary and tertiary structure prediction can be used to determine structures that have yet to be experimentally solved. A common method of predicting secondary structure from sequence is through free energy minimization via the Nearest-Neighbor Model, which states that each base pair's stability in a duplex is directly related to the identity of the base pairs directly adjacent to it. In this model, non-nearest neighbors are not taken into account. Through optical melting studies of GNRA tetraloops with differing non-nearest neighbors, this assumption was shown to be incorrect and that non-nearest neighbors do have a significant impact on the stability of the tetraloop. Using the results from these optical melting studies, an updated predictive model was created to better represent the impact of non-nearest neighbors on the thermodynamic stability of GNRA tetraloops. Since it has been shown that non-nearest neighbors do affect the thermodynamics of secondary structure motifs, such as bulge loops, internal loops, and hairpins, it is important to analyze these effects in the three-dimensional structures that have already been solved in order to better understand the role of non-nearest neighbors in structure and to find correlations between the thermodynamics and stability. The RNA Characterization of Secondary Structure Motifs (RNA CoSSMos) online database was created to be a freely available online resource to efficiently determine three-dimensional structure characteristics of common secondary structure motifs. Over 2,200 PDB files have been analyzed, annotated, and searched for false positives, and the resulting data has been put into a user-friendly, easily accessible online database. With this new online tool, the PDB was searched for GNRA tetraloops, and the structural information was collected. The data was sorted based upon the identities of the non-nearest neighbors, and the structural information was analyzed to find correlations between the identities of the non-nearest neighbors, the three-dimensional structures of the hairpin loops, and the thermodynamic contributions of these loops. |
