ChiRA: an integrated framework for chimeric read analysis from RNA-RNA interactome and RNA structurome data

Area of Science:

• Genomics
• Bioinformatics
• Molecular Biology

Background:

• Next-generation sequencing enables genome-wide RNA-RNA interaction and structure prediction.
• Chimeric reads from these experiments often map ambiguously due to short read lengths, complicating origin inference.
• Existing methods struggle with multi-mapped reads, limiting the discovery of RNA interactions.

Purpose of the Study:

• To present ChiRA, a novel framework for sensitive annotation of chimeric RNA reads.
• To improve the prediction of sequenced RNA hybrids and RNA-RNA interactions.
• To provide a comprehensive tool suite for RNA interactome and structurome analysis.

Main Methods:

• Developed ChiRA, a generic framework for chimeric read annotation.
• Implemented a strategy of grouping reference loci based on common reads and quantification.
• Integrated visualization tools (ChiRAViz) for intuitive dataset filtering.

Main Results:

• ChiRA improved correct read alignments up to 3-fold compared to common strategies.
• Identified shared protein families and pathways among genes at common read loci.
• Detected 3 times more new RNA-RNA interactions in published datasets than existing approaches.

Conclusions:

• ChiRA offers a sensitive and comprehensive solution for analyzing RNA-RNA interactome and structurome data.
• The framework effectively handles multi-mapped reads on paralogous genes without prior gene relation information.
• ChiRA facilitates the discovery of novel RNA-RNA interactions and provides robust visualization tools.