[Paper Review] Predicting complex 3D RNA structrues via a high resolution coarse-grained model

Tristan Cragnolini, Yoann Laurin+2 more|arXiv (Cornell University)|Apr 2, 2014

RNA and protein synthesis mechanismsBiochemistry, Genetics and Molecular Biology1 citations

TL;DR

HiRE-RNA is a high-resolution coarse-grained RNA model that accurately predicts equilibrium structures, dynamics, and thermodynamics of complex 3D RNA architectures by incorporating detailed base-pairing, stacking, and non-canonical interactions. The model successfully captures folding, stability, and free energy landscapes across 16 diverse RNA systems (12–76 nt), including monomers and dimers, over 850 μs of simulation time.

ABSTRACT

HiRE-RNA is a simplified, coarse-grained RNA model for the prediction of equilibrium configurations, dynamics and thermodynamics. Using a reduced set of particles and detailed interactions accounting for base-pairing and stacking we show that non-canonical and multiple base interactions are necessary to capture the full physical behavior of complex RNAs. In this paper we give a full account of the model and we present results on the folding, stability and free energy surfaces of 16 systems with 12 to 76 nucleotides of increasingly complex architectures, ranging from monomers to dimers, using a total of 850$\mu$s simulation time.

Motivation & Objective

• To develop a computationally efficient yet physically accurate RNA model capable of simulating complex 3D RNA structures.
• To investigate the role of non-canonical and multiple base interactions in determining RNA folding and stability.
• To predict equilibrium configurations, dynamics, and thermodynamics of structurally diverse RNAs, including dimers and complex architectures.
• To provide a scalable framework for studying RNA free energy surfaces and conformational landscapes.

Proposed method

• HiRE-RNA employs a reduced particle representation with explicit modeling of nucleotide-level interactions, including base pairing and stacking.
• The model incorporates detailed non-canonical and multiple base interactions to improve structural accuracy.
• A coarse-grained force field is designed to balance computational efficiency with physical realism in RNA folding simulations.
• Simulations are performed over 850 μs of total time across 16 RNA systems with varying complexity (12–76 nt), including monomers and dimers.
• Free energy surfaces are computed from simulation trajectories to analyze folding pathways and stability.

Experimental results

Research questions

• RQ1 How well can a high-resolution coarse-grained model predict the 3D structures of complex RNAs with non-canonical interactions?
• RQ2 What is the contribution of non-canonical and multiple base interactions to RNA folding and stability?
• RQ3 How do free energy landscapes of complex RNAs evolve across different structural architectures?
• RQ4 Can the model accurately simulate dynamics and equilibrium configurations across diverse RNA systems?

Key findings

• HiRE-RNA successfully predicts equilibrium structures of 16 RNA systems, including complex dimers and multimers, with high structural fidelity.
• Non-canonical and multiple base interactions are essential for capturing the full physical behavior of complex RNAs.
• The model achieves accurate representation of folding pathways and conformational dynamics over 850 μs of simulation time.
• Free energy surfaces computed from simulations reveal distinct folding landscapes, supporting the model's ability to predict thermodynamic stability.