GSE122713 Processing Pipeline

Publication

Immunity (2020) — PMID 32433949

Dataset

scRNA-seq, bulk RNA-seq, and ATAC-seq data from progenitor exhausted and terminally exhausted CD8+ T cells from tumors and chronic viral infection

1. 1

   Sample demultiplexing, barcode processing, alignment, filtering, UMI counting, and aggregation of multiple sequencing runs were performed using the Cell Ranger analysis pipeline (v1.2)

   Cell Ranger v1.2

   $ Bash example

   # Install Cell Ranger (example: download and add to PATH)
   # # Download Cell Ranger v1.2 from 10x Genomics website (requires login)
   # # e.g., wget https://cf.10xgenomics.com/releases/cell-exp/cellranger-1.2.0.tar.gz
   # # tar -xzf cellranger-1.2.0.tar.gz
   # # export PATH=/path/to/cellranger-1.2.0:$PATH
   
   # Define variables
   OUTPUT_ID="my_cellranger_analysis"
   FASTQ_DIR="/path/to/your/fastq_files" # Directory containing FASTQ files (e.g., from cellranger mkfastq)
   SAMPLE_NAME="my_sample"
   # Reference transcriptome (e.g., human GRCh38-2020-A)
   # Download from 10x Genomics: https://www.10xgenomics.com/resources/cell-ranger-downloads
   TRANSCRIPTOME_REF="/path/to/refdata-gex-GRCh38-2020-A"
   
   # Run Cell Ranger count to perform demultiplexing, barcode processing, alignment, filtering, and UMI counting
   # This command processes a single sample, potentially from multiple sequencing runs/lanes if FASTQs are in the specified directory.
   cellranger count \
       --id="${OUTPUT_ID}" \
       --transcriptome="${TRANSCRIPTOME_REF}" \
       --fastqs="${FASTQ_DIR}" \
       --sample="${SAMPLE_NAME}" \
       --localcores=8 \
       --localmem=64
   
   # If "aggregation of multiple sequencing runs" refers to combining outputs from multiple *samples*,
   # you would use cellranger aggr after running cellranger count for each sample.
   # Example for cellranger aggr:
   # # Create an aggregation CSV file (e.g., aggregation.csv) with sample_id and molecule_h5 paths
   # # sample_id,molecule_h5
   # # sample1,/path/to/sample1_output/outs/molecule_info.h5
   # # sample2,/path/to/sample2_output/outs/molecule_info.h5
   # # cellranger aggr --id="aggregated_output" --csv="aggregation.csv" --normalize=none
   

   Copy
2. 2

   Data normalization, dimension reduction, and differential expression analysis were performed using the Seurat R package (v2.2.0)

   Seurat v2.2.0 GitHub

   $ Bash example

   # Install R if not already present
   # sudo apt-get update
   # sudo apt-get install -y r-base
   
   # Install devtools if not already present
   # R -e 'install.packages("devtools")'
   
   # Install Seurat v2.2.0 using devtools::install_version
   # R -e 'devtools::install_version("Seurat", version = "2.2.0", repos = "http://cran.us.r-project.org")'
   
   # This script assumes Seurat v2.2.0 is installed and R is available.
   # It demonstrates the typical steps for normalization, dimension reduction, and differential expression.
   # Replace the dummy data loading with your actual input data (e.g., a count matrix).
   
   Rscript -e '
   library(Seurat)
   
   # --- 1. Load Data ---
   # In a real scenario, you would load your single-cell RNA-seq count matrix.
   # Example: pbmc.data <- Read10X(data.dir = "path/to/10x/data/")
   # For demonstration, we create a small dummy count matrix.
   set.seed(123)
   dummy_counts <- matrix(sample(0:100, 1000, replace = TRUE), nrow = 100, ncol = 10)
   rownames(dummy_counts) <- paste0("gene", 1:100)
   colnames(dummy_counts) <- paste0("cell", 1:10)
   pbmc <- CreateSeuratObject(raw.data = dummy_counts)
   
   # --- 2. Normalization ---
   # Normalizes the raw expression data. LogNormalize is the default for Seurat v2.
   pbmc <- NormalizeData(object = pbmc, normalization.method = "LogNormalize", scale.factor = 10000)
   
   # --- 3. Find Variable Genes (Optional but common before dimension reduction) ---
   # Identifies genes that are highly variable across cells.
   pbmc <- FindVariableGenes(object = pbmc, mean.function = ExpMean, dispersion.function = LogVMR, 
                             x.low.cutoff = 0.0125, x.high.cutoff = 3, y.cutoff = 0.5)
   
   # --- 4. Scale Data ---
   # Scales and centers the data. This is crucial before performing PCA.
   # Common covariates like "nUMI" (number of unique molecular identifiers) can be regressed out.
   pbmc <- ScaleData(object = pbmc, vars.to.regress = c("nUMI"))
   
   # --- 5. Dimension Reduction (PCA) ---
   # Performs Principal Component Analysis on the scaled data.
   pbmc <- RunPCA(object = pbmc, pc.genes = pbmc@var.genes, do.print = FALSE, pcs.print = 1:5, genes.print = 5)
   
   # --- 6. Cluster Cells (Optional but common after dimension reduction) ---
   # Finds clusters of cells based on their PCA embeddings.
   pbmc <- FindClusters(object = pbmc, reduction.type = "pca", dims.use = 1:10, resolution = 0.6, print.output = 0, save.SNN = TRUE)
   
   # --- 7. Differential Expression Analysis ---
   # Identifies genes that are differentially expressed between cell clusters.
   # Example: Find markers for cluster 0 compared to all other cells.
   # Replace "0" with the actual cluster ID you want to analyze.
   cluster_0_markers <- FindMarkers(object = pbmc, ident.1 = 0, min.pct = 0.25)
   
   # To find markers for all clusters, uncomment the following:
   # all_markers <- FindAllMarkers(object = pbmc, only.pos = TRUE, min.pct = 0.25, thresh.use = 0.25)
   
   # --- 8. Save Results (Optional) ---
   # Save the processed Seurat object and/or differential expression results.
   # saveRDS(pbmc, file = "processed_seurat_object.rds")
   # write.csv(cluster_0_markers, "cluster_0_markers.csv")
   # write.csv(all_markers, "all_cluster_markers.csv")
   '

   Copy View on GitHub
3. 3

   Single cell signatures scoring was performed using the FastProject software (v1.1.0)

   FastProject v1.1.0 GitHub

   $ Bash example

   # Install FastProject (if not already installed)
   # pip install fastproject
   
   # Example command for single cell signatures scoring with FastProject v1.1.0
   # This is a generic example as specific parameters (input files, output directory, etc.) are not provided in the description.
   # Replace 'input_expression_matrix.h5ad' with your actual single-cell expression data (e.g., anndata object).
   # Replace 'gene_signatures.gmt' with your actual gene signature file (e.g., in GMT format).
   # Replace 'output_directory' with your desired output location.
   fastproject \
     --expression_matrix input_expression_matrix.h5ad \
     --signatures gene_signatures.gmt \
     --output_dir output_directory

   Copy View on GitHub