scale expression by row or column in FeatureStatPlot plot_type = "heatmap"

[guillemsanchezsanchez1996]

Hello in FeatureStatPlot with plot_type = "heatmap" option is there a way to scale the expression data according to row or column?

Thanks for this amazing package :)

Guillem

[pwwang]

Hi, scaling expression by row/column in heatmap by FeatureStatPlot is not supported directly.

You can use layer_fun_callback to tweak it:

data(pancreas_sub)
features <- c(
   "Sox9", "Anxa2", "Bicc1", # Ductal
   "Neurog3", "Hes6", # EPs
   "Fev", "Neurod1", # Pre-endocrine
   "Rbp4", "Pyy", # Endocrine
   "Ins1", "Gcg", "Sst", "Ghrl" # Beta, Alpha, Delta, Epsilon
)

# generate a row/column-scaled expression value matrix
scaled_exprs <- ...

FeatureStatPlot(pancreas_sub, features = features, ident = "SubCellType",
   plot_type = "heatmap", name = "Expression Level", layer_fun_callback = function(j, i, x, y, w, h, fill, sr, sc) {
        # i: row indexes
        # j: column indexes
        # get the values from scaled_exprs[i, j] and use grid::grid_rect() to draw tiles 
        # to override the tiles in each cell
   })

I am not arguing if it is a good idea to do so, but I think it might be better to split the rows/columns into different sub-plots.