Specific splice junction detection in single cells with SICILIAN.

Area of Science:

• Genomics
• Molecular Biology
• Bioinformatics

Background:

• Accurate splice junction identification is crucial for single-cell biology research.
• Current single-cell RNA sequencing (scRNA-seq) platforms, like 10x Chromium, lack precise splice junction calling capabilities.
• Understanding splicing patterns is key to unraveling cellular complexity and function.

Purpose of the Study:

• To introduce SICILIAN, a novel computational method for assigning statistical confidence to splice junctions.
• To enhance the precision of splice junction detection in both bulk and single-cell RNA sequencing data.
• To enable the discovery of novel splicing events and regulatory mechanisms in single cells.

Main Methods:

• Developed SICILIAN, a method that leverages spliced aligner output to provide confidence scores for splice junctions.
• Applied SICILIAN to simulated and real scRNA-seq datasets, as well as matched bulk and single-cell data.
• Validated SICILIAN's performance by assessing accuracy, concordance, and agreement between replicates.

Main Results:

• SICILIAN demonstrates high accuracy in splice junction detection on simulated data.
• The method improves concordance between matched single-cell and bulk RNA sequencing datasets.
• SICILIAN increases the agreement between biological replicates in scRNA-seq experiments.
• SICILIAN successfully identifies previously unannotated splicing events in single cells.

Conclusions:

• SICILIAN significantly improves the precision of splice junction calls in scRNA-seq data.
• This enhanced precision facilitates more accurate biological interpretation of single-cell transcriptomes.
• SICILIAN opens new avenues for discovering novel splicing regulation and understanding cellular heterogeneity.