All-at-once RNA folding with 3D motif prediction framed by evolutionary information

Area of Science:

• Structural biology
• Computational biology
• Bioinformatics

Background:

• RNA molecules fold into complex three-dimensional (3D) structures crucial for their function.
• These structures are stabilized by canonical Watson-Crick base pairs and non-canonical interactions forming recurrent 3D motifs.
• Predicting these intricate structures remains a significant challenge in molecular biology.

Purpose of the Study:

• To develop a novel computational method for predicting RNA 3D structures.
• To jointly predict RNA secondary structure and 3D motifs from sequence or alignment data.
• To provide a comprehensive tool for analyzing RNA tertiary structures.

Main Methods:

• Introduction of CaCoFold-R3D, a probabilistic grammar for RNA structure prediction.
• Utilizing evolutionary information from RNA alignments to identify canonical helices and pseudoknots via covariation.
• Incorporating R3D grammars to model the spatial arrangement of RNA 3D motifs constrained by helix covariation.

Main Results:

• CaCoFold-R3D successfully predicts RNA 3D motifs and secondary structures simultaneously.
• The method accounts for over fifty known RNA motifs, predicting their occurrence in various loop regions.
• Demonstrated ability to arrange all structural elements into a single, unified RNA 3D structure prediction.

Conclusions:

• CaCoFold-R3D is a robust and accurate tool for predicting all-residue interactions in RNA 3D structures.
• The method is computationally efficient and adaptable for discovering novel RNA structural motifs.
• Presents a significant advancement in the field of RNA structure prediction and analysis.