samplePaths: Creates a set of sample path p-values for each length given a...

In NetPathMiner: NetPathMiner for Biological Network Construction, Path Mining and Visualization

Description

Randomly traverses paths of increasing lengths within a set network to create an empirical pathway distribution for more accurate determination of path significance.

Usage

samplePaths(graph, max.path.length, num.samples = 1000,
  num.warmup = 10, verbose = TRUE)

Arguments

graph	A weighted igraph object. Weights must be in edge.weights or weight edge attributes.
max.path.length	The maxmimum path length.
num.samples	The numner of paths to sample
num.warmup	The number of warm up paths to sample.
verbose	Whether to display the progress of the function.

Details

Can take a bit of time.

Value

A matrix where each row is a path length and each column is the number of paths sampled.

Author(s)

Timothy Hancock

Ahmed Mohamed

See Also

Other Path ranking methods: extractPathNetwork, getPathsAsEIDs, pathRanker

Examples

	## Prepare a weighted reaction network.
	## Conver a metabolic network to a reaction network.
 data(ex_sbml) # bipartite metabolic network of Carbohydrate metabolism.
 rgraph <- makeReactionNetwork(ex_sbml, simplify=TRUE)

	## Assign edge weights based on Affymetrix attributes and microarray dataset.
 # Calculate Pearson's correlation.
	data(ex_microarray)	# Part of ALL dataset.
	rgraph <- assignEdgeWeights(microarray = ex_microarray, graph = rgraph,
		weight.method = "cor", use.attr="miriam.uniprot",
		y=factor(colnames(ex_microarray)), bootstrap = FALSE)

	## Get significantly correlated paths using "p-valvue" method.
	##   First, establish path score distribution by calling "samplePaths"
 pathsample <- samplePaths(rgraph, max.path.length=10,
                        num.samples=100, num.warmup=10)

	##   Get all significant paths with p<0.1
	significant.p <- pathRanker(rgraph, method = "pvalue",
                sampledpaths = pathsample ,alpha=0.1)

NetPathMiner documentation built on Nov. 8, 2020, 8:20 p.m.