Skip to main content

scRNAseq Analysis in R with Seurat

5 Cluster Markers

Why do we need to do this?

Single cell data helps to segragate cell types. Use markers to identify cell types. warning: In this example the cell types/markers are well known and making this step easy, but in reality this step needs the experts curation.

5.1 Finding differentially expressed features (cluster biomarkers)

Tutorial: https://satijalab.org/seurat/articles/pbmc3k_tutorial#finding-differentially-expressed-features-cluster-biomarkers

5.2 Use markers to label or find a cluster

If you know markers for your cell types, use AddModuleScore to label them.

genes_markers <- list(Naive_CD4_T = c("IL7R", "CCR7"))

pbmc <- AddModuleScore(object = pbmc, features = genes_markers, ctrl = 5, name = "Naive_CD4_T",
    search = TRUE)


# notice the name of the cluster has a 1 at the end
names(pbmc@meta.data)
#> [1] "orig.ident"      "nCount_RNA"      "nFeature_RNA"   
#> [4] "percent.mt"      "RNA_snn_res.0.5" "seurat_clusters"
#> [7] "RNA_snn_res.0.1" "RNA_snn_res.2"   "Naive_CD4_T1"

# label that cell type
pbmc$cell_label = NA
pbmc$cell_label[pbmc$Naive_CD4_T1 > 1] = "Naive_CD4_T"
Idents(pbmc) = pbmc$cell_label

# plot
# Using a custom colour scale 
FeaturePlot(pbmc, features = "Naive_CD4_T1", label = TRUE, repel = TRUE, ) + scale_colour_gradientn(colours = c("lightblue","beige","red"))
#> Scale for colour is already present.
#> Adding another scale for colour, which will replace the
#> existing scale.