Described herein are systems and processes for designing the sequence of one or more interacting nucleic acid strands intended to adopt a target secondary structure at equilibrium. The target secondary structure is decomposed into a binary tree and candidate mutations are evaluated on leaf nodes of the tree. During a process of leaf optimization, defect-weighted mutation sampling is used to select each candidate mutation position with a probability proportional to its contribution to an ensemble defect of the leaf. Subsequences of the tree are then merged, moving up the tree until a final nucleotide sequence of interest is determined that has the target secondary structure at equilibrium.