Figure 4From: Shape and secondary structure prediction for ncRNAs including pseudoknots based on linear SVMAn example of structure selection based on hierarchical clustering. For each structure S j i in the folding space of sequence X i , the grammar string encoding the structure and the sequence is denoted as g s j i . All sequences and their associated structures are converted into grammar strings before clustering. The highlighted rectangles indicate grammar strings that are selected as representative structures.Back to article page