Networks

Networks

Networks can be displayed to represent the relevant associations between X and Y variables, they are inferred using a pairwise association score between X and Y variables. The user can set a threshold to represent only the variables X and Y with an association score greater than the threshold. By changing this threshold, the user can choose to include or exclude relationships in the relevance network. The output is a graph where each X- and Y-variable corresponds to a node and the edges included in the graph portray associations between them.

Usage in mixOmics

Networks graphs can be obtained in mixOmics via the function network as displayed below:

library(mixOmics)
## Will work with sPLS result
data(liver.toxicity)

    X <- liver.toxicity$gene
    Y <- liver.toxicity$clinic
    toxicity.spls <- spls(X, Y, ncomp = 3)

 network(toxicity.spls,
         color.node = c("orange","blue"),
         threshold = 0.75,
         row.names = FALSE, 
         col.names = FALSE)
network(toxicity.spls,color.edge = c("red","orange","cyan","blue"),row.names=FALSE, threshold = 0.75)
 network(toxicity.spls,lty.edge='dotted',row.names = FALSE, threshold = 0.75)

We can export the graph to Cytoscape file format using write.graph from the igraph package.

net.result <- network(toxicity.spls)

write.graph(net.result$gR, file = "network.gml", format = "gml")

References

González I., Lê Cao K.-A., Davis, M.D. and Déjean S. (2013) Insightful graphical outputs to explore relationships between two ‘omics’ data sets. BioData Mining 5:19