scINSIGHT for interpreting single-cell gene expression from biologically heterogeneous data.

Area of Science:

• Computational Biology
• Genomics
• Bioinformatics

Background:

• Single-cell RNA sequencing (scRNA-seq) generates vast datasets requiring integrative analysis.
• Existing batch effect removal methods struggle with heterogeneous scRNA-seq data across multiple biological conditions.

Purpose of the Study:

• To develop a novel method, scINSIGHT, for analyzing multi-condition scRNA-seq data.
• To learn coordinated gene expression patterns common or specific to different biological conditions.
• To identify cellular identities and processes across diverse single-cell samples.

Main Methods:

• scINSIGHT employs a novel approach to learn coordinated gene expression patterns.
• The method was validated using both simulated and real-world scRNA-seq datasets.
• Performance was compared against state-of-the-art computational methods.

Main Results:

• scINSIGHT demonstrated superior performance compared to existing methods.
• The method successfully identified common and condition-specific gene expression patterns.
• Cellular identities and biological processes were effectively discerned across samples.

Conclusions:

• scINSIGHT provides an effective solution for integrative analysis of heterogeneous scRNA-seq data.
• The method enhances the interpretation of biological similarities and differences across conditions.
• scINSIGHT is applicable to a wide range of biomedical and clinical research problems.