Processing of scRNAseq data

Illumina sequencer base call files (BCL) were processed and demultiplexed with CellRanger (v2.2) that used Illumina conversion software bcl2fastq2 (v2.2). FASTQ files were processed using CellRanger software, which performed alignment, filtering, barcode counting, and UMI counting. Single-cell V(D)J sequences were generated with VDJ function of CellRanger using a custom-made VDJ reference genome. For single-cell expression profiling, reads were aligned to the mouse genome (mm10-2.2), with CellRanger counts software, for their assignment as murine sequences. Low-quality cells were excluded in an initial quality control (QC) step by removing genes expressed in fewer than three cells. Also, cells with less than 200 genes expressed or that express >2500 genes were removed. We also removed cells that had >0.05% of mitochondrial-associated genes among their expressed genes. We then normalized the data using log normalization and scale factor 10,000 using the Normalize Data function in Seurat. To identify the differentially expressed features we used the Find Markers function with min.pct argument set to 0.25, which filters out genes expressed in <25% of the cells. PCA, t-SNE and graph clustering were produced using Cell Ranger’s pipeline with default settings. Analysis of DE genes downstream was performed using cLoupe (10× Genomics) and R package Seurat software⁵⁰.