Improved indel detection in DNA and RNA via realignment with ABRA2

Area of Science:

• Genomics
• Bioinformatics
• Computational Biology

Background:

• Genomic variant detection is crucial for diagnosing cancer and Mendelian disorders.
• Insertions and deletions (indels) are significant variations impacting gene function.
• Current indel detection methods, especially from RNA-Seq data, require improvement.

Purpose of the Study:

• To introduce ABRA2, an improved version of the ABRA tool for short-read realignment.
• To enhance the accuracy and scalability of indel detection from both RNA and DNA sequencing data.
• To broaden the applicability of variant detection to diverse sequencing types, including whole genomes.

Main Methods:

• ABRA2 is a redesigned implementation supporting realignment of RNA and DNA short reads.
• The tool is implemented using Java and C/C++.
• It offers improved scalability for analyzing human whole genomes.

Main Results:

• ABRA2 demonstrates substantial improvements in indel detection across various data types.
• The tool shows enhanced accuracy for whole genomes, targeted exomes, and transcriptome sequencing.
• It provides broader variant calling accuracy improvements compared to the original ABRA.

Conclusions:

• ABRA2 offers a significant advancement in indel detection for both DNA and RNA sequencing.
• The tool enhances variant calling accuracy across multiple genomic and transcriptomic analyses.
• ABRA2 provides a scalable and accurate solution for genomic variant detection in research and diagnostics.