simData: Simulate data.

In cbg-ethz/mnem: Mixture Nested Effects Models

Simulate data.

Description

This function simulates single cell data from a random mixture of networks.

Usage

simData(
  Sgenes = 5,
  Egenes = 1,
  Nems = 2,
  reps = NULL,
  mw = NULL,
  evolution = FALSE,
  nCells = 1000,
  uninform = 0,
  unitheta = FALSE,
  edgeprob = c(0, 1),
  multi = FALSE,
  subsample = 1,
  scalefree = FALSE,
  badCells = 0,
  exactProb = TRUE,
  tree = FALSE,
  ...
)

Arguments

Sgenes	number of Sgenes
Egenes	number of Egenes
Nems	number of components
reps	number of replicates, if set (not realistic for cells)
mw	mixture weights (has to be vector of length Nems)
evolution	evolving and not purely random network, if set to TRUE
nCells	number of cells
uninform	number of uninformative Egenes
unitheta	uniform theta, if TRUE
edgeprob	edge probability, value between 0 and 1 for sparse or dense networks or a range c(l,u) with lower and upper bound
multi	a vector with the percentages of cell with multiple perturbations, e.g. c(0.2,0.1,0) for 20 no quadruple knock-downs
subsample	range to subsample data. 1 means the full simulated data is used
scalefree	if TRUE, graph is scale free
badCells	number of cells, which are just noise and not connected to the ground truth network
exactProb	logical; if TRUE generates random network with exact fraction of edges provided by edgeprob
tree	if TRUE, restricts dag to a tree
...	additional parameters for the scale free network sampler (see 'nem' package)

Value

simulation object with meta information and data

Nem	list of adjacency matrixes generatign the data
theta	E-gene attachaments
data	data matrix
index	index for which Nem generated which cell (data column)
mw	vector of input mixture weights

Author(s)

Martin Pirkl

Examples

sim <- simData(Sgenes = 3, Egenes = 2, Nems = 2, mw = c(0.4,0.6))

cbg-ethz/mnem documentation built on Feb. 5, 2024, 5:46 a.m.