Abstract:Accurate free energy estimation is essential for the prediction of RNA secondary structure. The widely Turner model works well for nested structures. For pseudoknotted RNAs, there is no effective free energy rules. We present a new free energy estimation method that goes beyond the Turner model based on the physical consideration that the free energy of loops (including pseudoknotted loops) are mostly determined by the loop length and the end-end distance. Given a specific secondary structure (refer to as topology hereafter), we grow the RNA in three dimensional space and search among all possible arrangement of helices for stable structure. The free energy contribution of helices is estimated from the Turner model, and the free energy of loops is calculated based on our new approach. We assess our free energy model by testing its ability in selecting native structure from decoy topologies. For a testing set of 50-pseudoknotted RNAs, we create thousands of competitive decoy topologies and calculate their free energy. On average, our energy model provides an average of 7-8 percent improvement in term of sensitivity and specificity. In a blind pseudoknot structure prediction without inserting the native topology into the candidate list, our prediction has an average of 0.79 sensitives and 0.75 specificity. This performance is comparable to that of TdFOLD (0.76, 0.79) and better than Hotknots (0.69, 0.72), ILM(0.73, 0.69) and PKNOTS (0.78,0.73). Our free energy estimation method is fundamentally different from the others: it is based on the growing of conformations in three dimensional space and is well suited for predicting pseudoknotted structures. In addition, our method also gives coarse 3D conformation of a given secondary structure, which can lead to accurate 3D ab initio predicted structure upon further refinement.

Improvements of the Hierarchical Approach for Predicting Rna Tertiary Structure

Toward a Next-Generation Atlas of RNA Secondary Structure

A Hopfield Neural Network Based Algorithm for RNA Secondary Structure Prediction

Automated and Fast Building of Three-Dimensional RNA Structures

Ab initio Prediction of RNA Nucleotide Interactions with Backbone k-Tree Model

Methods and Applications of RNA Contact Prediction*

Prediction of the RNA Tertiary Structure Based on a Random Sampling Strategy and Parallel Mechanism

Evaluation of RNA secondary structure prediction for both base-pairing and topology

Advances in RNA-protein structure prediction

Optimization Of Rna 3d Structure Prediction Using Evolutionary Restraints Of Nucleotide-Nucleotide Interactions From Direct Coupling Analysis

Computational Prediction of RNA Tertiary Structures Using Machine Learning Methods*

Predicting 3D structure, flexibility and stability of RNA hairpins in monovalent and divalent ion solutions

A Local Structural Prediction Algorithm for RNA Triple Helix Structure

An RNA Base Discrete State Model Toward Tertiary Structure Prediction

RNA Fragment Modeling with a Nucleobase Discrete-State Model

Simultaneous prediction of RNA secondary structure and helix coaxial stacking

Systematic benchmarking of deep-learning methods for tertiary RNA structure prediction

Novel Representation of RNA Secondary Structure Used to Improve Prediction Algorithms.

Structure Prediction of RNA Loops with a Probabilistic Approach

Integrating end-to-end learning with deep geometrical potentials for ab initio RNA structure prediction

A New Free Energy Rule Toward Psuedo-knotted RNA Structure Prediction