getNetwork: Build Network

In ramellose/CoNetinR: CoNet implementation in R

Description

Build a network using Pearson, Spearman, Kullback-Leibler or Bray-Curtis

Usage

getNetwork(mat = matrix(), method = "spearman", T.up = 0.2,
  T.down = -0.2, shuffle.samples = F, norm = FALSE, rarefy = 0,
  stand.rows = FALSE, pval.cor = F, permut = F, renorm = F,
  permutandboot = F, iters = 100, bh = F, min.occ = 0,
  keep.filtered = F, plot = F, report.full = F, verbose = F)

Arguments

mat	user-provided matrix, can be left empty to generate one
method	network construction method, values: "pearson", "spearman", "kld" or "bray"
T.up	upper threshold for scores
T.down	lower threshold for scores
shuffle.samples	shuffle each sample before network construction
norm	normalize matrix before network construction
rarefy	if TRUE, rarefy matrix to the minimum total read count, if larger than 1, rarefy matrix to the given total read count
stand.rows	standardize rows by dividing each entry by its corresponding row sum, applied after normalization and rarefaction and before network construction
pval.cor	compute p-values of correlations with cor.test (cannot be selected together with permut, renorm and/or permutandboot or a method that is not a correlation)
permut	compute p-values on edges with a simple permutation test
renorm	compute p-values with a permutation test, using renormalization
permutandboot	compute p-values from both permutation and bootstrap distribution
iters	number of iterations for the permutation test
bh	multiple-test-correct using Benjamini-Hochberg
min.occ	only keep rows with at least the given number of non-zero values (carried out before network construction)
keep.filtered	sum all filtered rows and add the sum vector as additional row
plot	plot score or, if permut is true, p-value distribution
report.full	in addition to positive edge percentage, return other output variables
verbose	print the number of positive and negative edges and details of p-value computation

Details

Takes care of all steps of network construction, by also performing rarefication, normalization, permutation and bootstrapping. If ReNorm and permutandboot are both set to TRUE, this is equal to the ReBoot procedure implemented in CoNet. The original CoNet implementation is available at: http://psbweb05.psb.ugent.be/conet/

Value

postive edge percentage and more statistics depending on report.full

ramellose/CoNetinR documentation built on Oct. 16, 2019, 9:50 a.m.