Immune Single-Cell Annotation by Molecular Signature-Based Deconvolution with MIXTURE.

Area of Science:

• Genomics
• Immunology
• Bioinformatics

Background:

• Accurate cell type identification in single-cell RNA sequencing (scRNA-seq) is vital for dissecting the tumor immune microenvironment (TIME).
• Unsupervised annotation methods present limitations in detailed cell type analysis.
• Optimizing immunotherapy strategies relies on precise understanding of immune cell populations.

Purpose of the Study:

• To present a novel annotation protocol for scRNA-seq data based on the MIXTURE v-SVR deconvolution algorithm.
• To address the limitations of existing unsupervised cell type annotation methods.
• To enhance the granular identification of distinct cell types within complex scRNA-seq datasets.

Main Methods:

• Development of the MIXTURE v-SVR deconvolution algorithm for cell type annotation.
• Application of the MIXTURE protocol to an annotated melanoma scRNA-seq dataset.
• Analysis of unannotated clusters to explore composition and heterogeneity.

Main Results:

• The MIXTURE protocol effectively enhanced detailed cell type annotation in scRNA-seq data.
• Demonstrated improved composition analysis of unannotated clusters.
• Provided detailed insights into cluster heterogeneity, enabling more granular cell type identification.

Conclusions:

• The MIXTURE protocol offers a robust solution for accurate cell type identification in scRNA-seq datasets.
• This method significantly improves the analysis of tumor immune microenvironment composition.
• Enhanced cell type resolution facilitates more effective optimization of immunotherapy strategies.