Vertical nitrogen base stacking is a significant stabilizing interaction of DNA and RNA 3-D structures, which plays a major role in their folding and complexation. Stacking occurs more frequently between adjacent, but also non-adjacent, nucleotides, mostly in double-stranded helical regions. The stabilization of base stacking involves London dispersion forces , and interactions between partial charges within the adjacent rings . Evidences for hydrophobic forces between bases in solution , as well as a contradictory nonclassical hydrophobic effect , have been observed. However, these interactions were not characterized and parameterized such that they could define precise energy parameters that could be used for the detection of base stacking . Instead, a geometrical approach was chosen based on the method proposed by Gabb et al. .
We use relaxed ranges of the values defined in the Gabb et al. method to detect base stacking, even if somewhat large deviations from ideal parameters are observed. It has been shown that there are many inconsistencies in the atomic coordinates of RNA structures. The deviations measured in NMR spectroscopy and x-ray diffraction structures can be due to variations in the refinement protocols and force fields, as well as to artifacts resulting from the determination processes [3,28].
Therefore, we consider stacking between two nitrogen bases if the distance between their rings is less than 5.5Å, the angle between the two normals to the base planes is inferior to 30, and the angle between the normal of one base plane and the vector between the center of the rings from the two bases is less than 40. The class of a stacking interaction is defined by the nucleotides involved in it (Table 3.2).