Adenosine derived from CD39-expressing B cells promotes dysregulation of immune responses after sepsis [Scripts]

Summary methodology:

We re-analyzed single-cell transcriptomic data from the peripheral blood mononuclear cells (PBMC) of septic patients from a public dataset (Reyes et al., 2020). Basically, the samples were combined into two major groups for further comparisons, septic patients (sepsis) and non-septic patients (no sepsis). The sepsis group included samples from patients with urosepsis and patients with sepsis admitted to the medical intensive care unit, whereas the no sepsis group included samples from healthy patients and patients without sepsis admitted to the medical intensive care unit. B cells were filtered based on the authors' previous annotation for downstream analysis (Reyes et al., 2020). Specifically, the data matrices of 7970 B cells were imported to Seurat v3.1 (Stuart et al., 2019) by filtering genes expressed in at least 10 cells and more than 100 unique molecular identifiers (UMI) counts per cell. For the pre-processing step, outlier cells were filtered out based on three metrics (library size < 10000, number of expressed genes between 200 and 2000, and mitochondrial percentage expression < 5) resulting in a matrix with 13077 genes and 5104 cells. The remaining counts were normalized using the ‘LogNormalize’ method with a scale factor of 10000. The top 2,000 variable genes were then identified using the ‘vst’ method using the FindVariableFeatures function. Percent of mitochondrial genes was regressed out in the scaling step and Principal Component Analysis (PCA) was performed using the top 2,000 variable genes with 50 dimensions. Then, a dimensionality reduction method (UMAP) was performed on the top 7 PCAs. Additionally, a clustering analysis was performed on the first 7 principal components using a resolution of 0.6 followed by a dimensionality reduction method (UMAP). Then, differential gene expression analysis was performed using FindAllMarkers function to obtain a list of significant gene markers for each cluster of cells with default parameters. The plasmablast were identified by expression of CD39 gene count > 0 in the RNA essay (CD39) and BS3 plamablast’s markers previously described by Reyes et al. 2020. For enrichment analyses, we utilized the EnrichR tool (Chen et al., 2013) with the Reactome 2016 database for gene markers of plasmablasts.

Here we provide R scripts utilized for  single cell data analysis of B cell subpopulations: 

b_cells_sepsis_processing.R : Script to process the B cell population.

SEP_Bcells_sel.rds : RDS file generated by script b_cells_sepsis_processing.R

count_cell_frequencies_expressing_cd39.R: Script to count cell frequencies in cd39+ clusters.