Network functions and visualization

by Alexander Pico

2021-11-01

Source: vignettes/Network-functions-and-visualization.Rmd

This vignette will show how to perform basic network operations on an iGraph networks and use this information to customize its appearance in Cytoscape directly from R using the RCy3 package

*From Vessy’s “Fun with R blog”: http://www.vesnam.com/Rblog/viznets5/

Installation

if(!"RCy3" %in% installed.packages()){
    install.packages("BiocManager")
    BiocManager::install("RCy3")
}
library(RCy3)
if(!"igraph" %in% installed.packages()){
    install.packages("igraph")
}
library(igraph)
if(!"plyr" %in% installed.packages()){
    install.packages("plyr")
}
library(plyr)

Required Software

The whole point of RCy3 is to connect with Cytoscape. You will need to install and launch Cytoscape:

Read a data set.

Data format: dataframe with 3 variables; variables 1 & 2 correspond to interactions; variable 3 is weight of interaction

lesmis <- system.file("extdata","lesmis.txt", package="RCy3")
dataSet <- read.table(lesmis, header = FALSE, sep = "\t")

Create a graph. Use simplify to ensure that there are no duplicated edges or self loops

gD <- igraph::simplify(igraph::graph.data.frame(dataSet, directed=FALSE))

Verify the number of nodes (77) and edges (254):

igraph::vcount(gD)
igraph::ecount(gD)

Common iGraph functions

Calculate some node properties and node similarities that will be used to illustrate different plotting abilities

Calculate degree for all nodes

degAll <- igraph::degree(gD, v = igraph::V(gD), mode = "all")

Calculate betweenness for all nodes

betAll <- igraph::betweenness(gD, v = igraph::V(gD), directed = FALSE) / (((igraph::vcount(gD) - 1) * (igraph::vcount(gD)-2)) / 2)
betAll.norm <- (betAll - min(betAll))/(max(betAll) - min(betAll))
rm(betAll)

Calculate Dice similarities between all pairs of nodes

dsAll <- igraph::similarity.dice(gD, vids = igraph::V(gD), mode = "all")

Add attributes to network

Add new node attributes based on the calculated node properties/similarities

gD <- igraph::set.vertex.attribute(gD, "degree", index = igraph::V(gD), value = degAll)
gD <- igraph::set.vertex.attribute(gD, "betweenness", index = igraph::V(gD), value = betAll.norm)

Check the attributes. You should see “degree” and “betweeness” now, in addition to “name”.

And now for the edge attributes…

F1 <- function(x) {data.frame(V4 = dsAll[which(igraph::V(gD)$name == as.character(x$V1)), which(igraph::V(gD)$name == as.character(x$V2))])}
dataSet.ext <- plyr::ddply(dataSet, .variables=c("V1", "V2", "V3"), function(x) data.frame(F1(x)))

gD <- igraph::set.edge.attribute(gD, "weight", index = igraph::E(gD), value = 0)
gD <- igraph::set.edge.attribute(gD, "similarity", index = igraph::E(gD), value = 0)

Note: The order of interactions in dataSet.ext is not the same as it is in dataSet or as it is in the edge list and for that reason these values cannot be assigned directly

for (i in 1:nrow(dataSet.ext))
{
    igraph::E(gD)[as.character(dataSet.ext$V1) %--% as.character(dataSet.ext$V2)]$weight <- as.numeric(dataSet.ext$V3)
    igraph::E(gD)[as.character(dataSet.ext$V1) %--% as.character(dataSet.ext$V2)]$similarity <- as.numeric(dataSet.ext$V4)
}
rm(dataSet,dsAll, i, F1)

Check the edge attributes. You should see “weight” and “similarity” added.

Let check it out in Cytoscape

Update: You can go straight from igraph to Cytoscape, sending all attributes and displaying graph!

createNetworkFromIgraph(gD,new.title='Les Miserables')

Let’s decide on a layout

A list of available layouts can be accessed from R as follows:

We’ll select the “fruchterman-rheingold” layout. To see properties for the given layout, use:

getLayoutPropertyNames("fruchterman-rheingold")

We can choose any property we want and provide them as a space-delimited string:

layoutNetwork('fruchterman-rheingold gravity_multiplier=1 nIterations=10')

But that is a crazy layout, so let’s try “force-directed” instead:

layoutNetwork('force-directed defaultSpringLength=70 defaultSpringCoefficient=0.000003')

Next, we can visualize our data

On nodes…

setNodeColorMapping('degree', c(min(degAll), mean(degAll), max(degAll)), c('#F5EDDD', '#F59777', '#F55333'))
lockNodeDimensions(TRUE)
setNodeSizeMapping('betweenness', c(min(betAll.norm), mean(betAll.norm), max(betAll.norm)), c(30, 60, 100))

…and edges

setEdgeLineWidthMapping('weight', c(min(as.numeric(dataSet.ext$V3)), mean(as.numeric(dataSet.ext$V3)), max(as.numeric(dataSet.ext$V3))), c(1,3,5))
setEdgeColorMapping('weight', c(min(as.numeric(dataSet.ext$V3)), mean(as.numeric(dataSet.ext$V3)), max(as.numeric(dataSet.ext$V3))), c('#BBEE00', '#77AA00', '#558800'))

We will define our own default color/size schema after we defined node and edge rules, due to possible issues when using rules

Voila! All done.

Track versions for your records