Skip to main content
Fig. 1 | Genome Biology

Fig. 1

From: DotAligner: identification and clustering of RNA structure motifs

Fig. 1

Schematic of a pairwise alignment with DotAligner. A dynamic programming matrix is first filled in based on the similarity in sequence and cumulative base-wise pairing probabilities of two RNA sequences (top left: colour intensity indicates cumulative similarity score). A partition function over all pairwise alignments is calculated and interrogated for structural compatibility by stochastic backtracking (not illustrated). Two ensembles over all secondary structures are considered for this purpose (bottom left and top right dot plots: blue lines indicate cumulative base-wise pairing probabilities). The final scoring uses the base pair probabilities in the dot plots. This effectively warps the optimal sequence alignment path (top left, black outline) towards one that includes structural features (top left, blue outline and fill). In the bottom right, the optimal sequence alignment and associated consensus secondary structure is contrasted to that produced by DotAligner, exposing the common structural features hidden in the suboptimal base pairing ensemble of both sequences

Back to article page