R3D-BLAST2: an improved search tool for similar RNA 3D substructures

Area of Science:

• Bioinformatics
• Structural Biology
• Computational Biology

Background:

• RNA molecules perform diverse cellular functions determined by their 3D structures.
• The Protein Data Bank (PDB) hosts numerous RNA 3D structures.
• Tools for searching similar RNA structures are crucial for understanding structure-function relationships.

Purpose of the Study:

• To upgrade the R3D-BLAST tool for enhanced RNA 3D structure searching.
• To develop R3D-BLAST2, a web server with improved accuracy and functionality.
• To facilitate the identification of similar RNA 3D substructures.

Main Methods:

• Developed R3D-BLAST2 based on a higher-quality structural alphabet (SA).
• Encoded RNA 3D structures into 1D structural sequences using the SA.
• Modified the program to accept RNA structures in the mmCIF format.
• Applied BLAST to structural sequences for similarity searching.

Main Results:

• R3D-BLAST2 demonstrates superior performance compared to its predecessor, R3D-BLAST.
• The new tool outperforms other RNA structure search tools like RNA FRABASE, FASTR3D, and RAG-3D.
• R3D-BLAST2 successfully identified a larger number of similar RNA 3D substructures.

Conclusions:

• R3D-BLAST2 is a valuable, accurate BLAST-like tool for searching RNA 3D substructures in the PDB.
• The enhanced tool improves the discovery of structurally similar RNA motifs.
• R3D-BLAST2 is publicly accessible for research use.