do_LoadingsHeatmap() | PCA loadings

Loadings heatmaps show which genes contribute most to principal components. Displays top and bottom scoring genes for each PC alongside their expression levels.

Basic usage

p <- SCpubr::do_LoadingsHeatmap(sample = sample)
p

Select specific PCs

p <- SCpubr::do_LoadingsHeatmap(sample = sample,
                                dims = 1:2)
p

Adjust number of genes

p <- SCpubr::do_LoadingsHeatmap(sample = sample,
                                top_loadings = 2)
p

Group by cell type

p <- SCpubr::do_LoadingsHeatmap(sample = sample,
                                group.by = "subtype")
p

Subsample cells

p <- SCpubr::do_LoadingsHeatmap(sample = sample,
                                subsample = 1000)
p

Adjust color scales

p <- SCpubr::do_LoadingsHeatmap(sample = sample,
                                min.cutoff.loadings = -0.1,
                                max.cutoff.loadings = 0.1,
                                min.cutoff.expression = 0,
                                max.cutoff.expression = 3)
p

Parameter reference

Note

For parameters shared across many functions (color palettes, typography, legend styling, grid), see Shared features.

Core parameters

Parameter Description Default
dims PCs to include 1:10
top_loadings Top/bottom genes per PC 5
subsample Number of cells to sample NA

Scale limits

Parameter Description Default
min.cutoff.loadings Min loading score NA (q=0.05)
max.cutoff.loadings Max loading score NA (q=0.95)
min.cutoff.expression Min expression NA (0)
max.cutoff.expression Max expression NA (q=0.95)

See also