To build a network chart, your data must be formatted in one of these formats:

Adjacency Matrix. A square matrix, individuals in rows and columns are the same. Example: a correlation matrix

Incidence Matrix. Individuals in row and columns are not the same. Can be useful to check the releationship between one pop and another

Edge list. Edges are simply listed one by one.

Literal list. Connection are listed in a vector

A-B-C-D, E-A-E-A

1- Adjacency matrix

An adjacency matrix is a square matrix where individuals are the same in row and columns of the matrix. It’s typically the kind of matrix you get when calculating the correlation between each pair of individual. In this example, we have 1 connection from E to C, and 2 connections from C to E. By default, we get an unweighted and oriented network.

#library

library(igraph)

set.seed(10)

# Create data

data=matrix(sample(0:2, 25, replace=TRUE), nrow=5)

colnames(data)=rownames(data)=LETTERS[1:5]

# Tell Igraph it is an adjency matrix... with default parameters

set.seed(10)

network=graph_from_adjacency_matrix(data)

# plot it

plot(network)

Note that you can ask for weighted or unweighted network, and directed or undirected network. Following your choice, you need to specify how you want to interpret the adjency matrix: e.g. how to calculate the weight? Using the sum of the connection? The max? .. Type help(graph_from_adjacency_matrix) for more details.

# left

par(mfrow=c(1,2))

set.seed(10)

network=graph_from_adjacency_matrix(data, weighted=NULL)

plot(network, main="UNweighted")

# right

set.seed(10)

network=graph_from_adjacency_matrix(data, weighted=TRUE)

plot(network, main="weighted")

2- Incidence matrix

An incidence matrix does not necessarily have the same individuals in row and colum. By default, it is directed from rows to columns.

# create data

library(igraph)

set.seed(1)

data=matrix(sample(0:2, 15, replace=TRUE), nrow=3)

colnames(data) <- letters[1:5]

rownames(data) <- LETTERS[1:3]

# create the network object

set.seed(1)

network=graph_from_incidence_matrix(data)

# plot it

plot(network)

3- Edge List

The edge list is a data.frame listing all the connections.

# create data:

links=data.frame(

source=c("A","A", "A", "A", "A","F", "B"),

target=c("B","B", "C", "D", "F","A","E")

)

# create the network object

set.seed(10)

network=graph_from_data_frame(d=links, directed=F)

# plot it

plot(network)

You can add a second data frame which provides some features concerning each nodes. This kind of additional data can be useful to custom the network. Here the size of nodes depends of the “carac” column value. Note that you can make the chart directed or undirected.