Visualizing Cluster-specific Genes from Single-cell Transcriptomics Data Using Association Plots

Area of Science:

• Bioinformatics
• Computational Biology
• Genomics

Background:

• Visualizing single-cell transcriptomics data is challenging.
• Current methods often replace visualization with gene lists for cluster-specific expression analysis.

Purpose of the Study:

• To adapt Association Plots for visualizing cluster-specific genes in large single-cell datasets.
• To provide a user-friendly R package (APL) for this purpose.

Main Methods:

• Adapted Association Plots, a method from correspondence analysis, for single-cell data.
• Developed the APL R package.
• Applied APL to Peripheral Blood Mononuclear Cell data and human fetal gene expression data.

Main Results:

• Demonstrated delineation of novel marker genes using Association Plots.
• Showcased cell cluster annotation to known cell types.
• Compared Association Plots with differential gene expression methods and integrated with Gene Ontology Enrichment.

Conclusions:

• Association Plots offer an effective visualization strategy for cluster-specific genes in single-cell transcriptomics.
• The APL package facilitates marker gene discovery and cell type annotation.
• This approach enhances the interpretability of single-cell data analysis.