scPAFA

scPAFA enables rapid pathway activity score computation and identification of disease-related multicellular pathway modules in large-scale single-cell RNA sequencing datasets.

Key Features:

• Rapid PAS computation: Achieves >33-fold speedup in pathway activity score computation compared to existing tools for large-scale scRNA-seq datasets.
• Multicellular pathway module identification: Uncovers biologically interpretable disease-related multicellular pathway modules as low-dimensional representations of disease-related PAS variance across multiple cell types.

Scientific Applications:

• Colorectal cancer analysis: Identified reliable and interpretable multicellular pathway modules that capture transcriptomic features of tumor status in a dataset of 371,223 cells.
• Lupus atlas analysis: Identified transcriptional abnormalities in a large-scale lupus atlas of over 1.2 million cells.

Methodology:

scPAFA performs factor analysis on pathway activity scores to derive low-dimensional representations of disease-related variance across multiple cell types.