3dRNA: 3D Structure Prediction from Linear to Circular RNAs.

3D structures of RNAs are the basis for understanding their biological functions. However, experimentally solved RNA 3D structures are very limited. Therefore, many computational methods have been proposed to solve this problem, including our 3dRNA. 3dRNA is an automated template-based method of building RNA 3D structures from sequences and secondary structures by using the smallest secondary elements (SSEs) (http://biophy.hust.edu.cn/new/3dRNA). The first version of 3dRNA simply predicts an assembled structure for a target RNA. Later, it is improved to generate a set of assembled models and a method to further optimize them using experimental or theoretical restraints. In particular, pseudoknot base pairings are treated as restraints to solve the problem of no 3D templates for pseudoknots. Here 3dRNA is further extended to predict the 3D structures of circular RNAs since thousands of circular RNAs have been found recently but no 3D structures of them have been determined up to now. We show that circular RNAs can be divided into four types and two types show similar 3D structures with their linear counterparts while two types very different. We also show that the predicted structures of circular RNAs can bind to their ligands more stable than those of their linear counterparts, consistent with experimental results.