The RNA Folding Problems: Different Levels of sRNA Structure Prediction

Abstract

RNA 3D structure prediction is analogous to the protein folding problem, particularly the astronomical size of the conformational search space and the challenge of appropriately scoring native versus decoy alternatives. However, RNA presents important differences compared to proteins, notably the existence of a low-energy secondary structure intermediate on the pathway to tertiary folding. The availability of a secondary structure facilitates de novo prediction using assembly of fragments. RNA mutants and close homologs are readily predicted with high accuracy using homology modeling. Evolutionarily distant RNAs often require a combination of homology and de novo modeling approaches. The greatest challenges to RNA structure prediction are posed by multihelix loops, certain types of pseudoknots, and multidomain packing. There are also a variety of partial folding problems for RNA and opportunities for whole database structure prediction. Herein we describe a unified suite of programs called “RNA123” for the analysis and prediction of RNA structure.