vlnGrid: Generate a grid of tiny violins of gene expression in each...
In fungenomics/cytobox: A toolbox for analyzing single cell RNA-seq data

Description Usage Arguments Value Author(s) Examples

A grid of small violin plots, one plot per gene per cluster, similar to Figure 5D in the Drop-seq paper by Macosko et al. Aka "fish plot".

1 2	vlnGrid(seurat, genes, subset_clusters = NULL, order = "genes", colours = NULL, scale = "width", title = NULL, scales = "free_x")

`seurat`	Seurat object
`genes`	Genes to plot violins for, in the order (left to right) in which they should be plotted.
`subset_clusters`	(Optional) Vector of clusters to include on the plot.
`order`	Either "genes", or a vector containing the clusters in the order they should be plotted (from top to bottom). If "genes", clusters will be sorted in decreasing order of their median expression* of these genes, in the order in which they're provided. Passing a vector will cause clusters to be plot in that order. Default: "genes". *We take the median of cells with non-zero expression values (in `seurat@data`).
`colours`	(Optional) Vector of colours to use. Either one colour per cluster, in the order of `levels(seurat@ident)`, or one colour per cluster passed to `subset_clusters`, in the other they were provided. Default: default ggplot2 colours used by Seurat.
`scales`	See `scales` parameter in `facet_wrap`. Default: scales = "free_x", which allows every gene to have its own scale.
`width`	String, one of "width", "area", or "count". From the ggplot2 documentation of `geom_violin`: "if "area" (default), all violins have the same area (before trimming the tails). If "count", areas are scaled proportionally to the number of observations. If "width", all violins have the same maximum width." Default: "width".

A ggplot2 object

Selin Jessa

# Use the first 15 genes in the data, order by gene
vlnGrid(pbmc, head(rownames(pbmc@data), 15))

# Specify order
vlnGrid(pbmc, head(rownames(pbmc@data), 15), order = c(0, 3, 2, 1))

# Plot only a subset of clusters
vlnGrid(pbmc, head(rownames(pbmc@data), 15), subset_clusters = c(0, 1))

# Specify colours
vlnGrid(pbmc, head(rownames(pbmc@data), 15), subset_clusters = c(0, 1),
        colours = c("red", "blue"))