Physics-based all-atom modeling of RNA energetics and structure

Area of Science:

• Computational biology
• Biophysics
• Structural biology

Background:

• The exponential growth of RNA sequence data outpaces the understanding of RNA energetics, structures, and dynamics.
• All-atom computational methods, particularly molecular dynamics, offer potential solutions to bridge this knowledge gap.

Purpose of the Study:

• To review the current applications and limitations of molecular dynamics in RNA research.
• To provide a foundational understanding of the physical principles behind computational RNA methods for researchers.

Main Methods:

• Review of all-atom computational methods, focusing on molecular dynamics simulations for RNA.
• Discussion of advancements in algorithms and computing power enhancing prediction accuracy.
• Analysis of the requirements for benchmarking computational force fields through experimental data.

Main Results:

• Molecular dynamics can refine existing RNA structures from nuclear magnetic resonance and x-ray crystallography.
• Current methods are unreliable for predicting RNA structures solely from sequence information.
• Significant research is needed to understand RNA folding interactions and their quantitative approximation.

Conclusions:

• Computational methods are becoming more reliable and accessible, poised to become standard tools for biologists.
• Further development and extensive experimental validation are crucial for advancing sequence-based RNA structure prediction.
• This review serves as a guide for those initiating computational studies on RNA molecules.