JACUSA: site-specific identification of RNA editing events from replicate sequencing data

Area of Science:

• Molecular Biology
• Genomics
• Bioinformatics

Background:

• RNA editing, a co-transcriptional modification, diversifies RNA molecules and alters protein-coding sequences.
• Adenosine deaminases acting on RNA (ADAR) enzymes catalyze the common A→I RNA editing.
• Current methods for detecting RNA editing sites, such as RNA-DNA differences (RDDs), lack robustness for replicate RNA-seq data and rarely address condition-specific events (RNA-RNA differences, RRDs).

Purpose of the Study:

• To develop a robust computational tool for detecting single nucleotide variants from RNA-DNA and RNA-RNA comparisons.
• To provide a comprehensive solution for identifying both standard and condition-specific RNA editing events.
• To evaluate the performance of the developed tool against existing variant callers.

Main Methods:

• Development of JACUSA, a novel software solution for variant detection.
• In silico benchmarking of JACUSA against other variant callers using F-measure (precision and recall).
• Experimental validation using whole genome and RNA sequencing of HEK-293 cells and analysis of publicly available Drosophila melanogaster RNA-seq data.

Main Results:

• JACUSA demonstrates superior performance, particularly in RNA-RNA comparisons, achieving higher F-measures than existing algorithms.
• The software exhibits higher recall and comparable precision in detecting A→I editing sites from RNA-DNA comparisons in HEK-293 cells.
• JACUSA effectively captures a majority of A→I events from RNA-RNA comparisons across different species and cell lines.

Conclusions:

• JACUSA is a versatile tool for detecting single nucleotide variants from RNA-DNA and RNA-RNA sequencing data.
• The software offers improved accuracy, especially for identifying condition-specific RNA editing events.
• JACUSA provides a significant advancement for the robust detection of RNA editing modifications.