Detecting unannotated splicing events in short-read RNA-seq with SAMI, a UMI-aware Nextflow pipeline

Area of Science:

• Genomics
• Molecular Biology
• Bioinformatics

Background:

• RNA sequencing (RNA-seq) has largely replaced microarrays for gene expression profiling.
• The full potential of RNA-seq for clinical splicing analysis remains underutilized.
• Existing tools often target large cohorts or known isoforms, limiting their use in routine diagnostics for novel events.

Purpose of the Study:

• To introduce SAMI (Splicing Analysis with Molecular Indexes), a UMI-aware pipeline for detecting splicing events.
• To address the limitations of current tools in identifying non-recurring splicing events in low-dimension clinical datasets.
• To provide a user-friendly tool with graphical representations and tunable filtering for accurate splicing analysis.

Main Methods:

• Developed SAMI as a fully-integrated, UMI-aware pipeline.
• Utilized the STAR aligner as a foundation.
• Implemented novel post-processing steps for gaps and intron retentions to enhance accuracy.
• Assessed SAMI's performance on real and simulated data, including intragenic splicing aberrations and gene fusions.

Main Results:

• SAMI is designed to detect splicing events that deviate from standard transcript annotations.
• The pipeline offers clear graphical representations and adjustable filtering stringency.
• Performance was evaluated against concurrent software using commercial control samples and simulated data.

Conclusions:

• SAMI enhances the capability of RNA-seq for detailed splicing analysis in clinical settings.
• The tool is particularly valuable for identifying rare or non-recurring splicing events in routine diagnostics.
• SAMI offers an accurate and accessible solution for advancing clinical splicing analysis.