Relate Deconvolution Subgroups to Pathway Activities — compute.subgroup.pathways • multideconv

Correlates deconvolution subgroup profiles with a pre-computed pathway activity matrix, saves one heatmap per cell type to Results/ and returns the correlations and p-values.

Usage

compute.subgroup.pathways(
  subgroups,
  pathways = NULL,
  file_name = "Test",
  height = NULL,
  width = NULL,
  par_mar = c(4, 25, 5, 3),
  pval = 0.05,
  corr_type = "pearson"
)

Arguments

subgroups: Output list from compute.deconvolution.analysis().
pathways: A numeric matrix or data frame with samples as rows and pathway activities as columns. Row names must match sample identifiers in subgroups.
file_name: Character prefix used when naming output PDF files.
height, width: Plot height and width in inches (passed to ggplot2::ggsave()). If NULL (default), the size is chosen from the number of subgroups and pathways.
par_mar: Ignored; kept for backwards compatibility.
pval: P-value threshold; correlations above this are not starred.
corr_type: Correlation type, "pearson" (default) or "spearman".

Value

Invisibly, a list with one element per cell type, each holding correlations and pvalues (subgroups as rows, pathways as columns). One PDF heatmap per cell type is also saved in Results/.