Guidelines for single cell RNA sequencing analysis of eosinophils

Eosinophils are challenging to profile by single cell RNA sequencing (scRNA-seq) approaches due to their fragile nature and the abundance of RNases and cytotoxic enzymes stored in cytoplasmic granules, which can compromise RNA integrity upon stress. Although recent technical advances have improved eosinophil recovery, their transcriptomes remain intrinsically sparse, particularly in mature cells, resulting in low gene detection and high dropout that can bias standard preprocessing and quality-control steps.

Here, we integrated multiple publicly available eosinophil scRNA-seq datasets from our laboratory and other groups, and performed comparative analyses across platforms, tissues, and species. We show that eosinophils consistently display among the lowest transcriptome coverage, emphasizing the need for eosinophil-adapted analytical strategies. To enable reliable eosinophil annotation despite high dropout rates, we curated a dedicated eosinophil marker-gene panel derived from cross-dataset differential expression signatures. We further demonstrate that intron-inclusive genome alignment markedly increases eosinophil gene and transcript detection compared with exon-only alignment. Finally, we identify genotype-dependent programs: Il5-transgenic eosinophils exhibit a less mature profile, whereas wild-type eosinophils show stronger host-defense-associated signatures.

Together, these results provide a practical framework for eosinophil-focused scRNA-seq analysis that improves eosinophil recovery, annotation, and biological interpretation in complex datasets.

This repository contains additional processed files from GSE282765 and GSE182001 that were generated with the BD Rhapsody(TM) WTA Analysis Pipeline (v1.12.1) and gene code GRCm38 (v25) for mouse and GRCh38 (v43) for human data. 

Each zip folder contains one type of analysis: 

BD_automatic_cell_determination_intronic_and_exonic_reads: Automatic cell deterimination, gene mapping to intronic and exonic reads

BD_forced_cell_determination_exonic_reads_only: Forced cell determination, gene mapping to exonic reads only 

BD_forced_cell_determination_intronic_and_exonic_reads: Forced cell determination, gene mapping to intonic and exonic reads 

Sample_tag_calls_from_BD_forced_cell_determination_intronic_and_exonic_reads: Sample tag calls for each cell ID from forced cell determination, gene mapping to intronic and exonic reads

Unfiltered_counts_BD_forced_cell_determination_intronic_and_exonic_reads: Unfiltered counts from forced cell determination, gene mapping to intronic and exonic reads

The publication can be found here: https://doi.org/10.1093/jleuko/qiag055

The analysis code associated with this study can be found on the GitHub repository: https://github.com/Arnold-Lab-UZH/Guidelines_scRNAseq_analysis_eosinophils