IndepthPathway

IndepthPathway performs pathway enrichment analysis for scRNA-seq and bulk RNA sequencing by applying Weighted Concept Signature Enrichment Analysis (WCS-Enrichment) to mitigate noise and low gene coverage.

Key Features:

• WCS-Enrichment algorithm: Implements Weighted Concept Signature Enrichment Analysis (WCS-Enrichment) to assess functional relationships between pathway gene sets and differentially expressed genes.
• Universal concept signature: Constructs a cumulative "universal concept signature" from highly differentially expressed genes to mitigate scRNA-seq noise and low gene coverage.
• Single-cell focus: Targets scRNA-seq datasets and addresses challenges of high technical noise, low gene coverage, and detection in less abundant cell populations.
• Bulk RNA sequencing support: Extends applicability to bulk RNA sequencing datasets in addition to single-cell data.
• Simulation-based validation: Validated with simulations incorporating technical variability and gene expression dropouts typical of scRNA-seq data.
• Benchmarking on matched data: Benchmarked against matched single-cell and bulk RNA-seq datasets to assess performance.
• Stable enrichment results: Produces stable and deep pathway enrichment results despite stochasticity in input data.

Scientific Applications:

• Single-cell pathway enrichment: Identify enriched pathways and functional processes in scRNA-seq data, including in low-abundance cell populations.
• Bulk RNA-seq pathway analysis: Perform pathway enrichment analysis on bulk RNA sequencing datasets.
• Comparative analysis: Compare pathway-level signals between matched single-cell and bulk RNA-seq datasets.
• Method robustness assessment: Use simulation studies that model technical variability and gene expression dropouts to evaluate method sensitivity and stability.

Methodology:

Implements WCS-Enrichment using a cumulative "universal concept signature" derived from highly differentially expressed genes; validated via simulations incorporating technical variability and gene expression dropouts and by benchmarking against matched single-cell and bulk RNA-seq data.