|
Bioinformatics
Structural Bioinformatics of RNA refers to efforts to integrate the exponentially growing databases of RNA sequences with the database of high-resolution x-ray structures. The high-resolution 3D structures of representative large and small ribosomal subunits provide much of the impetus for these efforts (1-4). These structures have provided large-scale confirmation of the hierarchical and modular organization of RNA structure and the compact folding mediated by recurrent motifs (5). The basepair is the most basic recurrent motif. We have provided a comprehensive classification of basepairing based on the geometric concept of isostericity, which is key for understanding how RNA molecules evolve, with plasticity at the level of sequence but conservation of 3D structure (6,7). Isostericity Matrices were constructed which organize all canonical and non-Watson-Crick basepairs into isosteric subsets. This identifies those base substitutions which can occur within a motif while preserving its structure. This approach provides the basis for analyzing and classifying RNA motifs in complex structures such as the ribosome, and aids in the prediction of 3D structure based on sequence variation in homologous RNA molecules (8,9). It also provides the basis for structure-based alignment of homologous sequences. This is being extended to the analysis of base-stacking interactions. Computer programs for extracting and correlating key structural parameters for each nucleotide in a complex structure, including key torsion angles and local chain orientation, and for classifying stacking and basepairing interactions have been developed.
The following displays a few examples
of programs which have been developed.
Program output from base pair classification program, which displays the relative position of the interacting nucleotide.
Base pair covariation analysis program, which displays the isostericity within the geometric family.
Motif covariation analysis program.
References: