ASGAL: aligning RNA-Seq data to a splicing graph to detect novel alternative splicing events

Area of Science:

• Bioinformatics
• Computational Biology
• Genomics

Background:

• Transcript reconstruction from RNA-Seq is complex, but splicing event detection is feasible.
• Current methods align RNA-Seq reads to a reference genome and compare with gene annotations, often using splicing graphs.
• This approach is adequate for many transcriptome analyses.

Purpose of the Study:

• To present ASGAL (Alternative Splicing Graph ALigner), a tool for mapping RNA-Seq data to splicing graphs.
• To detect novel splicing events, including those with unannotated splice sites.
• To compute spliced alignments and identify novel events relative to gene annotations.

Main Methods:

• ASGAL maps RNA-Seq data directly to a splicing graph representation of gene annotations.
• It takes annotated transcripts and RNA-Seq samples as input.
• The tool computes spliced alignments for each read and lists novel splicing events.

Main Results:

• ASGAL successfully enriches gene annotations with novel alternative splicing events.
• It demonstrates higher accuracy in predicting novel splice site events compared to other tools.
• The tool performs well even when genes express a single isoform.

Conclusions:

• ASGAL is the first tool to detect novel alternative splicing events by direct alignment to splicing graphs.
• It offers improved accuracy and prediction of novel events compared to genome-based alignment methods.
• The tool enhances transcriptome analysis by identifying previously undiscovered splicing variations.