RNA Folding Based on 5 Beads Model and Multiscale Simulation.

Area of Science:

• Computational Biology
• Biophysics
• Molecular Modeling

Background:

• All-atom molecular dynamics simulations are computationally expensive and limited to small RNA molecules.
• Current coarse-grained models often rely on parameters derived from known RNA structures, hindering the study of modified RNAs.

Purpose of the Study:

• To develop a novel coarse-grained model (AIMS_RNA-B5) for enhanced RNA folding prediction.
• To improve the accuracy and efficiency of simulating modified RNA structures.

Main Methods:

• Proposed the AIMS_RNA-B5 model with a 5-bead representation (3 for bases, 2 for the backbone).
• Utilized all-atom molecular dynamics simulations (AAMDS) to generate trajectories for fitting coarse-grained force field (CGFF) parameters.
• Performed coarse-grained molecular dynamics simulations (CGMDS) for conformation sampling and accelerated folding.

Main Results:

• The AIMS_RNA-B5 model demonstrated improved performance compared to the AIMS_RNA-B3 model.
• Successfully simulated the folding of hairpin, pseudoknot, and tRNA structures.
• The AAMDS-guided CGMDS approach enhances conformational sampling and folding speed.

Conclusions:

• The AIMS_RNA-B5 model offers a more reasonable and effective approach for RNA folding prediction.
• This model facilitates the study of modified RNA structures, overcoming limitations of previous methods.