scLVM

scLVM models hidden sources of variation in single-cell RNA sequencing (RNA-seq) data to identify and correct confounding factors such as cell cycle effects, thereby improving detection of genuine gene expression heterogeneity and cellular subpopulations.

Key Features:

• Latent Variable Modeling: scLVM employs latent variable models to represent hidden factors that influence gene expression variability across single cells.
• Confounding Factor Correction: scLVM identifies and corrects for confounding sources of variation, explicitly including cell cycle effects, to refine expression signals.
• Unbiased Transcriptome Assays: scLVM is applied to unbiased single-cell transcriptome measurements (single-cell RNA-seq) from hundreds of individual cells to facilitate discovery of novel cellular subpopulations.

Scientific Applications:

• Identification of Subpopulations: scLVM enables identification of previously undetectable cellular subpopulations, exemplified by distinguishing stages in differentiation of naive T cells into T helper 2 cells.
• Gene Expression Heterogeneity Analysis: scLVM separates multiple sources of gene expression heterogeneity in single-cell transcriptomes to provide insight into cellular dynamics and function.

Methodology:

Uses computational latent variable models to capture and adjust for hidden confounding factors in single-cell RNA-seq data.