A unified analysis of cell-type- and trajectory-associated pathways in single-cell data using Phoenix

Area of Science:

• Single-cell genomics
• Computational biology
• Systems biology

Background:

• Single-cell RNA sequencing (scRNA-seq) enables detailed analysis of cellular heterogeneity.
• Identifying biological pathways driving cell types and trajectories from scRNA-seq data remains challenging.
• Existing methods struggle with subtle, nonlinear pathway activities and limited interpretability.

Purpose of the Study:

• To develop Phoenix, a novel pathway analysis framework for scRNA-seq data.
• To accurately identify biological pathways associated with cell types and cellular trajectories.
• To enhance biological interpretability and uncover complex gene interactions in single-cell data.

Main Methods:

• Phoenix utilizes random forest models and non-parametric significance testing.
• It evaluates functional gene sets for distinguishing cell types and ordering cells along pseudotemporal trajectories.
• The framework quantifies pathway effect sizes and identifies both up- and downregulated processes.

Main Results:

• Phoenix successfully identified cell type-specific and trajectory-associated pathways in hematopoiesis and embryogenesis datasets.
• The framework detected nonlinear pathway activities and complex gene interactions.
• Phoenix demonstrated superior performance in capturing small pathway activities and revealed cross-species pathway overlap.

Conclusions:

• Phoenix offers a sensitive and interpretable framework for pathway analysis in scRNA-seq data.
• It effectively uncovers biologically meaningful pathways and their component interactions.
• The tool facilitates deeper exploration of dynamic gene regulation in complex biological systems.