In [1]:
library(WGCNA)
library(igraph)
library("flashClust")
library('stringr')
options(stringsAsFactors = FALSE);
enableWGCNAThreads();
Reading the graph file¶
In [18]:
setwd("D:\\FIU\\CyFinder\\Data")
df <- read.delim("Breast_Cancer.txt",header=FALSE, check.names=FALSE)
In [19]:
df
Constructing the igraph graph object. Then plotting the graph and get the adjacency matrix from it¶
In [20]:
G <- graph.data.frame(d=df,directed=FALSE)
In [22]:
plot(G, vertex.label = V(G)$name)
In [31]:
adjMat <- as_adjacency_matrix(G, sparse=FALSE)
In [33]:
head(adjMat)
Calculating Tom Based Dissimilarity from the Adjacency Matrix¶
In [106]:
tomSimilarity <- TOMsimilarity(adjMat)
tomDist = 1-tomSimilarity
In [107]:
rownames(tomDist) <- rownames(adjMat)
colnames(tomDist) <- colnames(adjMat)
In [108]:
collectGarbage()
Do Average Linkage Hierarchical Clustering based on TOM based dissimilarity matrix, and plot the dendrogram¶
In [109]:
hierADJ=hclust(as.dist(tomDist), method="average")
In [110]:
par(mfrow=c(1,1))
plot(hierADJ, main="Clustering tree based of the module eigengenes")
module identification from the result of clustering algorithm, dendrogram¶
In [116]:
# We like large modules, so we set the minimum module size relatively high:
minModuleSize = 4;
# Module identification using dynamic tree cut:
unmergedLabels = cutreeDynamic(dendro = hierADJ, distM = tomDist,
deepSplit = 2, cutHeight = 0.21,
minClusterSize = minModuleSize,
pamRespectsDendro = FALSE );
unmergedColors = labels2colors(unmergedLabels)
In [117]:
unmergedColors
In [118]:
plotDendroAndColors(hierADJ, unmergedColors, "Dynamic Tree Cut",
dendroLabels = FALSE, hang = 0.03,
addGuide = TRUE, guideHang = 0.05)
Each cluster index and number of nodes in it¶
In [134]:
#first bold number is the index of the node
#2nd number is their cluster index
table(unmergedLabels)
color assignment of every nodes¶
In [140]:
for (i in 1:84){
print(paste(colnames(adjMat)[i],unmergedColors[i]))
}