ClusteredDotPlot: ClusteredDotPlot

In bimberlabinternal/CellMembrane: A package with high-level wrappers and pipelines for single-cell RNA-seq tools

ClusteredDotPlot

Description

A function that generates a clustered dot plot with a heatmap of scaled expression.

Usage

ClusteredDotPlot(
  seuratObj,
  features,
  groupFields = "ClusterNames_0.2",
  assay = "RNA",
  scaling = "column",
  layer = "data",
  dotSizeScaleFactor = 50,
  forceRescaling = FALSE,
  inferDefaultArguments = TRUE,
  printInferredArguments = FALSE,
  numberColumns = T,
  numberRows = T,
  ...
)

Arguments

seuratObj	The Seurat object that holds the data.
features	The features to plot.
groupFields	The metadata column that is used for grouping.
assay	The assay to plot.
scaling	The scaling method for the heatmap. Options are "row", "column", or none.
layer	The layer of the Seurat object that holds the relevant expression data.
dotSizeScaleFactor	An integer used to scale the dot size. Larger values result in smaler max dot size
forceRescaling	A boolean that determines if the Seurat object should be rescaled to include entries in the features vector if any are missing from the scale.data layer. This might be costly to perform locally.
inferDefaultArguments	If TRUE, the function will infer the default arguments for the ComplexHeatmap::Heatmap function.
printInferredArguments	Boolean to control optional printing of the arguments inferred by inferDefaultArguments.
numberColumns	Boolean controlling the behavior of column titling by inferDefaultArguments. If TRUE, this will label each column's K means clusters with numeric titles.
numberRows	Boolean controlling the behavior of row titling by inferDefaultArguments. If TRUE, this will label each row's K means clusters with numeric titles.
...	Additional arguments to pass to ComplexHeatmap::Heatmap

Examples

## Not run: 
#set the seurat Idents for FindAllMarkers
Seurat::Idents(seuratObj) <- "ClusterNames_0.2"

markers <- Seurat::FindAllMarkers(seuratObj)

#filter markers to display the largest cluster identity markers according to average log fold change and differences in pct expression.
strong_markers <- markers[abs(markers$avg_log2FC) > 3 & abs(markers$pct.1 - markers$pct.2) > 0.25, "gene"]

dotPlot <- ClusteredDotPlot(seuratObj, features = strong_markers, groupFields = "ClusterNames_0.2", scaling = 'column')
print(dotPlot)

#Additionally, you can establish logical groupings based on a mixture of metadata fields
 #create individual classifications based on lineage markers
 CD3E_positive <- Seurat::WhichCells(seuratObj, expression = CD3E > 0)
 CD20_positive <- Seurat::WhichCells(seuratObj, expression = MS4A1 > 0)
 CD14_positive <- Seurat::WhichCells(seuratObj, expression = CD14 > 0)
 seuratObj$CD3E_positive <- ifelse(rownames([e-mail]) %in% CD3E_positive, "CD3E_Pos", "CD3E_Neg")
 seuratObj$CD20_positive <- ifelse(rownames([e-mail]) %in% CD20_positive, "CD20_Pos", "CD20_Neg")
 seuratObj$CD14_positive <- ifelse(rownames([e-mail]) %in% CD14_positive, "CD14_Pos", "CD14_Neg")
 #roll lineage markers up into a single classification
 seuratObj$CellClassification <- paste0(seuratObj$CD3E_positive, "_", seuratObj$CD20_positive, "_" , seuratObj$CD14_positive)
 suppressWarnings(ClusteredDotPlot(seuratObj, 
                                 features = c("CD3E", "MS4A1", "CD14"),
                                 groupFields = c("CellClassification"),
                                 layer = 'data',
                                 scaling = "none", 
                                 height = grid::unit(10, "cm"), 
                                 width = grid::unit(4.5, "cm"), 
                                 show_row_dend = FALSE, 
                                 inferDefaultArguments = F
                               ))


## End(Not run)

bimberlabinternal/CellMembrane documentation built on Feb. 24, 2025, 4:30 a.m.