SVN_links: Statistical validated networks

In SVN: Statistically Validated Networks

Description

Statistical validated networks

Usage

SVN_links(states_vs_t, alpha = 0.01, MHT.correction = "bonferroni",
  exclude.states = NULL, states.pair.types = "all",
  alternative = "overexpression")

Arguments

states_vs_t	a matrix with time in lines and individual states in columns
alpha	the family-wise error rate in the case of Bonferroni multiple-hypothesis correction, or the false discovery rate in the case of the FDR multiple hypothesis correction
MHT.correction	the type of multiple hypothesis correction
exclude.states	a vector of states to remove from states_vs_t
states.pair.types	accepted value: "all", "same", "different": selects which kind of state pairs are tested: if there are two states e.g. (1,2), "all" allows (1,1), (2,2) and (1,2); "same" allows (1,1) and (2,2), while "different" allows (1,2)
alternative	a string either equal to "overexpression" or "underexpression"

Value

a data.table object of the node pairs significantly correlated at the alpha level. The columns are: link_id: a unique identifier for links; i and j are the node names; si and sj are their states; pv is the value associated to the link; threshold is the Multiple Hypothesis Testing-adjusted p-value threshold (only links with pv<=threshold are kept)

Examples

x=sample(c(1,0),1000,replace = TRUE)
xx=x; x[1]=1-x[1]   # one modifies just one element
x_rev=rev(x)
y=sample(c(1,0),1000,replace = TRUE)
M=cbind(x,xx,x_rev,y)
mylinks=SVN_links(M)
print(mylinks)
# one can compute clusters as well
myclusters=SVN_clusters(mylinks)

Example output

Loading required package: igraph

Attaching package: ‘igraph’

The following objects are masked from ‘package:stats’:

    decompose, spectrum

The following object is masked from ‘package:base’:

    union

Loading required package: memoise

SVN documentation built on Sept. 24, 2019, 9:05 a.m.