sc-ImmuCC: hierarchical annotation for immune cell types in single-cell RNA-seq

Area of Science:

• Immunology
• Bioinformatics
• Computational Biology

Background:

• Accurate immune cell identification in single-cell RNA sequencing (scRNA-Seq) data is essential for studying immune responses in health and disease.
• Challenges include data heterogeneity, sparsity, and gene expression similarity among immune cell types.

Purpose of the Study:

• To develop a novel tool, sc-ImmuCC, for hierarchical annotation of immune cell types from scRNA-Seq data.
• To improve the accuracy and comprehensiveness of immune cell identification in complex biological datasets.

Main Methods:

• Developed sc-ImmuCC, a tool utilizing optimized gene sets and the ssGSEA algorithm for hierarchical annotation.
• Implemented a three-layer hierarchical annotation strategy simulating natural immune cell differentiation.
• Validated performance across diverse tissue datasets.

Main Results:

• sc-ImmuCC achieved stable performance with average accuracy ranging from 71-90% across different tissue datasets.
• The tool successfully annotates nine major immune cell types and 29 cell subtypes.
• Application to COVID-19 and influenza datasets revealed a significantly higher proportion of monocytes in patients compared to healthy individuals.

Conclusions:

• sc-ImmuCC provides a robust and user-friendly method for comprehensive immune cell annotation in scRNA-Seq data.
• The optimized gene sets and hierarchical strategy can enhance other annotation methods.
• The tool facilitates the study of immune mechanisms in physiological and pathological conditions, including infectious diseases.