AEnet: a practical tool to construct the splicing-associated phenotype atlas at a single cell level

Area of Science:

• Genomics
• Molecular Biology
• Bioinformatics

Background:

• Alternative splicing (AS) drives proteomic diversity and cellular heterogeneity, impacting physiological and pathological processes.
• Single-cell RNA sequencing (scRNA-seq) captures AS events at cellular resolution but faces limitations like shallow depth and high dropout rates.
• Existing clustering methods often neglect the combined influence of AS and gene expression on cell type definition.

Purpose of the Study:

• To develop a novel method, AEnet, integrating gene expression and AS patterns for comprehensive cellular heterogeneity profiling.
• To define "cell subpopulations" by considering both AS and gene expression dynamics.
• To identify key AS events and their regulatory mechanisms driving cell state transitions.

Main Methods:

• Developed Alternative Splicing-Gene Expression Network (AEnet) to combine gene expression levels with AS patterns.
• Applied AEnet to profile cellular heterogeneity in tumor cells, pan-cancer immune cells, and embryonic cells.
• AEnet identifies key AS events and infers underlying regulatory mechanisms.

Main Results:

• AEnet demonstrated enhanced cell clustering compared to previous methods.
• Identified novel AS events with potential functional significance across different cell types.
• Discovered key splicing factors involved in cell state transitions in tumor, immune, and embryonic cells.

Conclusions:

• AEnet provides a powerful approach to dissect cellular heterogeneity by integrating AS and gene expression.
• The method offers new insights into the functional importance of AS in physiological and pathological contexts.
• AEnet facilitates the discovery of regulatory mechanisms governing cell state transitions.