EPIC

EPIC estimates cell type proportions from bulk gene expression data to quantify immune, cancer, and other nonmalignant cell-type composition in heterogeneous samples such as tumor specimens.

Key Features:

• RNA-seq-based reference profiles: EPIC uses RNA sequencing (RNA-seq)-derived reference gene expression profiles for immune cells and other nonmalignant cell types found in tumors.
• User-defined references: EPIC accepts user-supplied gene expression reference profiles for custom cell types.
• Handling uncharacterized cell types: EPIC accounts for uncharacterized cell types within samples to include components not represented in the reference panel.
• Renormalization for mRNA content: EPIC performs a renormalization step to adjust for differences in mRNA content across cell types.
• Single-cell RNA-seq integration: EPIC leverages single-cell RNA sequencing (scRNA-seq) data to derive or refine biologically relevant reference profiles.

Scientific Applications:

• Oncology research: Estimating immune, cancer, and other nonmalignant cell proportions in tumor specimens to study tumor microenvironment composition, disease progression, and treatment response.
• Confounding factor adjustment: Quantifying cellular composition to manage confounding effects in downstream bulk gene expression analyses.
• Precision medicine: Providing cell composition estimates to inform patient-specific assessments of tumor biology relevant to tailored treatment approaches.

Methodology:

EPIC integrates RNA-seq-derived reference gene expression profiles with single-cell RNA-seq data, applies normalization and a renormalization/adjustment for mRNA content, and computes cell-type proportion estimates from bulk gene expression data.