ELLIPSIS: robust quantification of splicing in scRNA-seq.

Area of Science:

• Genomics
• Computational Biology
• Molecular Biology

Background:

• Alternative splicing is a crucial, cell-type-specific regulatory process.
• Quantifying splicing in single-cell RNA sequencing (scRNA-seq) is difficult due to low and uneven data coverage.

Purpose of the Study:

• To develop a robust method for quantifying alternative splicing in scRNA-seq data.
• To enable the detection of differential splicing, including novel splicing events, particularly in cancer cells.

Main Methods:

• Developed ELLIPSIS, a computational tool for splicing quantification.
• Leveraged local read coverage, flow conservation, and intra-cell type similarity.
• Applied ELLIPSIS to simulated and glioblastoma scRNA-seq data.

Main Results:

• ELLIPSIS accurately estimates Percent Spliced In values from simulated data.
• The tool reliably detects differential splicing between cell types.
• Identified differentially spliced genes in glioblastoma, linked to cell migration and neuronal guidance.

Conclusions:

• ELLIPSIS provides a robust solution for splicing analysis in scRNA-seq.
• The identified genes offer insights into glioblastoma progression and cell behavior.