abc: ABC algorithm for network reverse-engineering

In networkABC: Network Reverse Engineering with Approximate Bayesian Computation

ABC algorithm for network reverse-engineering

Description

ABC algorithm for network reverse-engineering

Usage

abc(
  data,
  clust_coeffs = c(0.33, 0.66, 1),
  tolerance = NA,
  number_hubs = NA,
  iterations = 10,
  number_networks = 1000,
  hub_probs = NA,
  neighbour_probs = NA,
  is_probs = 1
)

Arguments

data	: Any microarray data in the form of a matrix (rows are genes and columns are time points)
clust_coeffs	: one dimensional array of size clust_size of clustering coefficients (these clustering coefficient are tested in the ABc algorithm).
tolerance	: a real value based for the tolerance between the generated networks and the reference network
number_hubs	: number of hubs in the network
iterations	: number of times to repeat ABC algorithm
number_networks	: number of generated networks in each iteration of the ABC algorithm
hub_probs	: one-dimensional array of size number_genes for the each label to be in the role of a hub
neighbour_probs	: this is the matrix of neighbour probabilities of size number_nodes*number_nodes
is_probs	: this needs to be set either to one (if you specify hub_probs and neighbour_probs) or to zero (if neither probabilities are specified). Warning: you should specify both hub_probs and neighbour_probs if is_probs is one. If is_prob is zero these arrays should simply indicate an array of a specified size..

Examples

M<-matrix(rnorm(30),10,3)
result<-abc(data=M)

networkABC documentation built on Oct. 19, 2022, 1:08 a.m.