create_bulks: simulate bulk data by summing over single-cell data

In MarianSchoen/DMC: Comparison of different Deconvolution Models in several scenarios

simulate bulk data by summing over single-cell data

Description

simulate bulk data by summing over single-cell data

Usage

create_bulks(
  exprs,
  pheno,
  cell.type.column = "cell_type",
  n.bulks = 500,
  include.in.bulks = NULL,
  n.profiles.per.bulk = 1000,
  sum.to.count = TRUE,
  frac = 5,
  switch.prob = 0
)

Arguments

exprs	non-negative numeric, scRNA-seq profiles as columns, features as rows
pheno	data.frame, phenotype data containing cell type labels for the expression matrix, must contain 'cell.type.column', ordering of cells must be the same as in exprs
cell.type.column	string, which column of 'pheno' holds the cell type information? default "cell_type"
n.bulks	integer, the number of bulks to be created, defaults to 500
include.in.bulks	vector of strings, cell types to be used for bulk simulation; if not supplied, all will be used
n.profiles.per.bulk	positive numeric, number of samples to be randomly drawn for each bulk; default 1000
sum.to.count	boolean, should all bulks be normalized to a fixed total count number (1e6)? default TRUE
frac	numeric >= 1; determines the variance of cell type proportions within the simulated bulks; default 5
switch.prob	numeric, 0 <= switch.prob <= 1, fraction of bulks sampled with inverse cell type quantities; default 0

Value

list with - "bulks": matrix containing bulk expression profiles (features x bulks) - "props" matrix containing quantities (cell type x bulks)

MarianSchoen/DMC documentation built on Aug. 2, 2022, 3:05 p.m.