RNA Sequencing-Based Immune Cell Deconvolution

Definition

RNA sequencing-based immune cell deconvolution is a computational method that estimates the proportion and types of immune cells present in bulk tissue RNA-seq data. It allows researchers to infer immune composition from gene expression profiles, without the need for single-cell or flow cytometry data.

• Works on bulk RNA-seq data, which includes mixed cell populations
• Uses known immune cell gene signatures to deconvolute expression
• Output is typically a cell type proportion matrix (e.g. % CD8+ T cells, % macrophages)

• CIBERSORT / CIBERSORTx – Reference-based method using a leukocyte signature matrix (LM22)
• xCell – Uses gene set enrichment to score cell types
• EPIC – Designed for tumor environments
• MCP-counter – Estimates abundance of immune and stromal populations
• TIMER – Focused on tumor immune estimation across cancer types

• Characterize the tumor immune microenvironment (TIME)
• Predict response to immunotherapy
• Stratify patients based on immune infiltration patterns
• Complement histological or flow-based findings

In lung adenocarcinoma RNA-seq data, CIBERSORT may reveal elevated M2 macrophages and reduced CD8+ T cells in non-responders to PD-1 blockade therapy.

• Accuracy depends on the quality of the reference signature
• Cannot capture spatial information
• Performance can vary with tumor heterogeneity or stromal contamination