CTD (version 0.99.8)

graph.diffuseP1: Diffuse Probability P1 from a starting node

Description

Recursively diffuse probability from a starting node based on the connectivity of the network, representing the likelihood that a variable is most influenced by a perturbation in the starting node.

Usage

graph.diffuseP1(p1,sn,G,vNodes,thresholdDiff,adj_mat,verbose=FALSE,
                        out_dir="",r_level=1,coords=NULL)

Arguments

p1

- The probability being dispersed from the starting node, sn, which is preferentially distributed between network nodes by the probability diffusion algorithm based solely on network connectivity.

sn

- "Start node", or the node most recently visited by the network walker, from which p1 gets dispersed.

G

- A list of probabilities, with names of the list being the node names in the network.

vNodes

- "Visited nodes", or the history of previous draws in the node ranking sequence.

thresholdDiff

- When the probability diffusion algorithm exchanges this amount (thresholdDiff) or less between nodes, the algorithm returns up the call stack.

adj_mat

- The adjacency matrix that encodes the edge weights for the network, G.

verbose

- If debugging or tracking a diffusion event, verbose=TRUE will activate print statements. Default is FALSE.

out_dir

- If specified, a image sequence will generate in the output directory specified.

r_level

- "Recursion level", or the current depth in the call stack caused by a recursive algorithm. Only relevant if out_dir is specified.

coords

- The x and y coordinates for each node in the network, to remain static between images. Only relevant if out_dir is specified.

Value

G - A list of returned probabilities after the diffusion of probability has truncated, with names of the list being the node names in the network.

Examples

# NOT RUN {
# Read in any network via its adjacency matrix
adj_mat=rbind(c(0,1,2,0,0,0,0,0,0), #A's neighbors
                c(1,0,3,0,0,0,0,0,0), #B's neighbors
                c(2,3,0,0,1,0,0,0,0), #C's neighbors
                c(0,0,0,0,0,0,1,1,0), #D's neighbors
                c(0,0,1,0,0,1,0,0,0), #E's neighbors
                c(0,0,0,0,1,0,0,0,0), #F's neighbors
                c(0,0,0,1,0,0,0,1,0), #G's neighbors
                c(0,0,0,1,0,0,1,0,0), #H's neighbors
                c(0,0,0,0,0,0,0,0,0) #I's neighbors
                )
rownames(adj_mat)=c("A","B","C","D","E","F","G","H","I")
colnames(adj_mat)=c("A","B","C","D","E","F","G","H","I")
G=vector(mode="list", length=ncol(adj_mat))
names(G)=colnames(adj_mat)
G=lapply(G, function(i) i[[1]]=0)
probs_afterCurrDraw=graph.diffuseP1(p1=1.0, sn=names(G)[1], G=G,
                                    vNodes=names(G)[1], 
                                    thresholdDiff=0.01, adj_mat, TRUE)
# }