RAG: an update to the RNA-As-Graphs resource

Area of Science:

• Bioinformatics
• Computational Biology
• Molecular Biology

Background:

• The RNA-As-Graphs (RAG) resource, initially developed in 2004, aids in the discovery of novel RNA structures.
• RAG previously classified, cataloged, and predicted RNA secondary structure motifs using clustering and build-up methods.

Purpose of the Study:

• To update the RAG resource with enhanced features for analyzing RNA structures.
• To improve the classification and prediction of RNA secondary structure motifs.
• To stimulate the search for novel RNA functionality and design.

Main Methods:

• Implemented a new supervised clustering algorithm to classify RNA motifs as existing, RNA-like, or non-RNA-like.
• Generated candidate RNA-like topologies using the algorithm with existing RNAs as training data.
• Enhanced RAG with expanded search capabilities, improved database interface, and refined graph analysis utilities.

Main Results:

• The supervised clustering algorithm demonstrated considerable improvements over previous methods.
• Identified 126 tree and 16,658 dual graphs as potential RNA-like topologies.
• The updated RAG resource offers advanced graph-based exploration and design suggestions for RNA motifs.

Conclusions:

• The updated RAG resource effectively augments the discovery of novel RNA functionality.
• It provides tools for classifying existing motifs, suggesting new ones for design, and enabling specific topology searches.
• RAG serves as a valuable graph-based tool for exploring RNA motifs and designing new RNA molecules.