Prediction of interacting single-stranded RNA bases by protein-binding patterns

Area of Science:

• Structural biology
• Computational biology
• Drug discovery

Background:

• Protein-RNA interactions are vital for gene regulation.
• Understanding atomic-level interactions aids in developing targeted therapies.
• Extruded nucleotides in RNA, not involved in base pairing, are key interaction sites.

Purpose of the Study:

• To develop a novel framework for predicting protein binding sites for RNA dinucleotides.
• To identify and classify shared 3-D physicochemical binding patterns.
• To enable the prediction of RNA fragments and aid in drug design.

Main Methods:

• Classification of known RNA nucleotide and dinucleotide protein binding sites.
• Spatial multiple alignment for identifying shared binding patterns.
• Utilizing predicted binding sites to construct novel RNA fragments.

Main Results:

• Identified common 3-D physicochemical binding patterns in protein binding sites.
• Revealed similarities between dinucleotide binding sites across diverse proteins.
• Successfully predicted RNA dinucleotide binding sites and interacting RNA fragments.
• Demonstrated a drug design application by identifying potential small-molecule scaffolds.

Conclusions:

• A novel framework accurately predicts RNA dinucleotide binding sites and interacting fragments based on protein structure.
• The findings facilitate the design of novel RNA fragments and offer new strategies for drug discovery targeting protein-RNA interactions.