AI Insight
EXTRARNAS is a new Java-based software framework that automates the extraction and standardization of RNA base-pairing interactions from three-dimensional structural data. The tool addresses the problem that existing RNA annotation tools produce inconsistent outputs, particularly for non-canonical base pairs, by processing multiple annotation tools simultaneously and converting their outputs into standardized formats including BPSEQ and a newly introduced BPSEQE format. Testing on eight RNA structures with triple-helix motifs revealed systematic differences between annotation tools, demonstrating the need for standardized comparison methods.
Why it matters
This framework enables researchers to systematically compare RNA structure annotations from different tools and obtain more reliable structural data for benchmarking computational predictions and analyzing RNA function. The standardized output formats facilitate reproducible large-scale RNA structural analysis and could improve the accuracy of data-driven RNA structure prediction methods.
Understand the Science
⚠️ Preprint – Noch nicht peer-reviewed
Dieser Artikel wurde noch nicht von unabhängigen Experten begutachtet. Die Ergebnisse sind vorläufig und sollten mit Vorsicht interpretiert werden.
Accurate annotation of RNA base-pairing interactions is essential for structural analysis, benchmarking, and data-driven RNA structure prediction. Several tools can extract RNA interactions from three-dimensional coordinates, but their outputs are heterogeneous and may disagree, particularly for non-canonical base pairs. We present EXTRARNAS, a Java-based framework for automated, reproducible, and user-friendly large-scale extraction of RNA structural annotations with multiple tools. EXTRARNAS processes batches of RNA structures specified by PDB identifier and chain, or provided as local PDB files, executes annotation tools through a Docker-based environment, and parses tool-specific outputs using ANTLR4-based grammars. For each structure-tool pair, the framework generates standard BPSEQ files for canonical cis Watson-Crick interactions and introduces BPSEQE, a standardized text format for representing the extended secondary structure, preserving canonical, non-canonical, and multiple interactions per nucleotide. The current prototype supports RNAView, MC-Annotate, and RNAPolis Annotator. We demonstrate EXTRARNAS on eight RNA structures containing triple-helix motifs, comparing extracted canonical pairs against curated BPSEQ references and evaluating the recovery of manually validated Hoogsteen interactions. The results show consistent differences among tools, especially for non-canonical interactions, highlighting the need for standardized representations such as BPSEQE to support reproducible comparison and future consensus-based annotation.
Source: EXTRARNAS: A Framework for Extracting RNA Structures with Multiple Tools