simBoolGtn: Sample random network and simulate data

In MartinFXP/B-NEM: Training of logical models from indirect measurements of perturbation experiments

Sample random network and simulate data

Description

Draws a random prior network, samples a ground truth from the full boolean extension and generates data

Usage

simBoolGtn(
  Sgenes = 10,
  maxEdges = 25,
  stimGenes = 2,
  layer = 1,
  frac = 0.1,
  maxInDeg = 2,
  dag = TRUE,
  maxSize = 2,
  maxStim = 2,
  maxInhibit = 1,
  Egenes = 10,
  flip = 0.33,
  reps = 1,
  keepsif = FALSE,
  negation = 0.25,
  allstim = FALSE,
  and = 0.25,
  positive = TRUE,
  verbose = FALSE
)

Arguments

Sgenes	number of S-genes
maxEdges	number of maximum edges (upper limit) in the DAG
stimGenes	number of stimulated S-genes
layer	scaling factor for the sampling of next Sgene layerof the prior. high (5-10) mean more depth and low (0-2) means more breadth
frac	fraction of hyper-edges in the ground truth (GTN)
maxInDeg	maximum number of incoming hyper-edges
dag	if TRUE, graph will be acyclic
maxSize	maximum number of S-genes in a hyper-edge
maxStim	maximum of stimulated S-genes in an experiment (=data samples)
maxInhibit	maximum number of inhibited S-genes in an experiment (=data samples)
Egenes	number of E-genes per S-gene, e.g. 10 S-genes and 10 E-genes will return 100 E-genes overall
flip	fraction of inhibited E-genes
reps	number of replicates
keepsif	if TRUE does not delete sif file, which encodes the prior network
negation	sample probability for negative or NOT edges
allstim	full network in which all S-genes are also stimulated
and	probability for AND-gates in the GTN
positive	if TRUE, sets all stimulation edges to activation, else samples inhibitory edges by 'negation' probability
verbose	TRUE for verbose output

Value

list with the corresponding prior graph, ground truth network and data

Author(s)

Martin Pirkl

Examples

sim <- simBoolGtn()
plot(sim)

MartinFXP/B-NEM documentation built on Oct. 27, 2023, 8:24 p.m.