Variables representation
See also this page for more details about correlation circles (2D and 3D plots).
## Only displays the names of the clinical variables plotVar(result, comp = 1:2, Y.label = TRUE) ## Displays both names plotVar(result, comp = 1:2, Y.label = TRUE, X.label = TRUE, cex = c(0.5, 0.8))
This graph only displays the genes and clinical variables that have been selected with sPLS. A thorough investigation would be worthwhile to examine the biological link between these clusters of genes with some of the clinical variables.
3D variables plot:
## Here we do not display the name of the genes plot3dVar(result, comp = 1:3, Y.label = TRUE, axes.box = "axes")
It is even easier to see the clusters in this representation if you rotate the image. The name of the genes can also be displayed.
Relevance networks, as well as Clustered Image Map (CIM), will facilitate the interpretation of the results:
## By setting keep.var = TRUE, we only display the variables selected by sPLS
## on dimensions 1 and 2
color.edge <- colorRampPalette(c("darkgreen", "green", "yellow", "red", "darkred"))
network(result, comp = 1:2, keep.var = TRUE, shape.node = c("rectangle", "rectangle"),
color.node = c("white", "pink"), color.edge = color.edge(10), alpha = 3)
This network can be saved as a .glm for an input into Cytoscape, see here.
The Clustered Image Map (see reference below) also allows to visualize the correlation between the variables:
## The variables selected on the chosen dimensions are displayed (here we chose ## the 3 dimensions) cim(result, comp = 1:3, xlab = "genes", ylab = "clinic", margins = c(5, 6))
References
- Bushel P., Wolfinger R. D. and Gibson G. (2007) Simultaneous clustering of gene expression data with clinical chemistry and pathological evaluations reveals phenotypic prototypes. BMC Systems Biology1(15).
- Lê Cao K.-A., Rossouw D., Robert-Granié C. and Besse P. (2008) A sparse PLS for variable selection when integrating Omics data. Statistical Applications in Genetics and Molecular Biology 7, article 35.
More on relevance networks and CIM:
- González I., Lê Cao K.-A., Davis M. D. and Déjean S. (2011) Insightful graphical outputs to explore relationships between two ‘omics’ data sets. Submitted.