“…Among the main reasons for the lag in long-range motif exploration are the limitations of existing computational methods designed for the motif search problem, which is formulated as the task of identifying instances of a given 3D motif within a given RNA 3D structure or a set of structures [11,[13][14][15][16][17][18]. The commonly used methods are limited by at least one of three major restraints: RNA sequence (types of bases) [14,15], annotations of interactions (e.g., types of base pairs [19]) [13,15,16,18,20], and backbone topology (types of loops [21,22]) [11,12,14,16,17,[23][24][25]. Sequence restraints may hide rare non-canonical motif variants, such as the U-minor variant of the A-minor interaction motif [26].…”