Fig. 4
From: Automated design of dynamic programming schemes for RNA folding with pseudoknots
The two types of canonical representations for the helices of a graph completion G and associated dynamic programming schemes. (Left) In the Diagonal case, only the sequence positions of external (resp. internal) anchors are provided. Internal ones are obtained as the base case of an energy model-dependent dedicated dynamic programming scheme, propagating values for anchors in S along the way. (Right) In the clique case, all four anchors delimiting the helix have known position. Again, a dedicated dynamic programming algorithm is used to optimize over all possible contents for the helix, while accounting for associated free-energies