dnet (version 1.1.7)

dNetConfidence: Function to append the confidence information from the source graphs into the target graph

Description

eConsensusGraph is supposed to append the confidence information (extracted from a list of the source graphs) into the target graph. The confidence information is about how often a node (or an edge) in the target graph that can be found in the input source graphs. The target graph is an object of class "igraph" or "graphNEL", and the source graphs are a list of objects of class "igraph" or "graphNEL". It also returns an object of class "igraph" or "graphNEL"; specifically, the same as the input target graph but appended with the "nodeConfidence" attribute to the nodes and the "edgeConfidence" attribute to the edges.

Usage

dNetConfidence(target, sources, plot = F)

Arguments

target

the target graph, an object of class "igraph" or "graphNEL"

sources

a list of the source graphs, each with an object of class "igraph" or "graphNEL". These source graphs will be used to calculate how often a node (or an edge) in the target graph that can be found with them.

plot

logical to indicate whether the returned graph (i.e. the target graph plus the confidence information on nodes and edges) should be plotted. If it sets true, the plot will display the returned graph with the size of nodes indicative of the node confidence (the frequency that a node appears in the source graphs), and with the width of edges indicative of the edge confidence (the frequency that an edge appears in the source graphs)

Value

an object of class "igraph" or "graphNEL", which is a target graph but appended with the "nodeConfidence" attribute to the nodes and the "edgeConfidence" attribute to the edges (in the form of 100 percentage)

See Also

visNet

Examples

# NOT RUN {
# 1) generate a target graph according to the ER model
g <- erdos.renyi.game(100, 1/100)
target <- dNetInduce(g, V(g), knn=0)

# 2) generate a list source graphs according to the ER model
sources <- lapply(1:100, function(x) erdos.renyi.game(100*runif(1),
1/10))

# 3) append the confidence information from the source graphs into the target graph
g <- dNetConfidence(target=target, sources=sources)

# 4) visualise the confidence target graph
visNet(g, vertex.size=V(g)$nodeConfidence/10,
edge.width=E(g)$edgeConfidence)
# }