CIM

Clustered Image Maps

…is based on a hierarchical clustering simultaneously applied on the rows and the columns of a real-valued similarity matrix. This similarity matrix is obtained via the results of (r)CCA or (s)PLS. The matrix is graphically represented as a 2-dimensional colored image, where each entry of the matrix is colored on the basis of its value, and where the rows and columns are reordered according to a hierarchical clustering. Dendrograms resulting from the clustering are added to the left (or right) side and to the top (or bottom) of the image.

Usage in mixOmics

CIM can be obtained in mixOmics via the function cim as displayed below:

## We consider the results of a rcc computation on the nutrimouse data
data(nutrimouse)
X <- nutrimouse$lipid
Y <- nutrimouse$gene
result <- rcc(X, Y, ncomp = 3, lambda1 = 0.064, lambda2 = 0.008)

## Margins is the values for column and row margins and zoom = TRUE provides
## an interactive zoom
cim(result, comp = 1:3, xlab = "genes", ylab = "lipids", margins = c(5, 6), zoom = TRUE)

R script: 

You should get two windows like these ones, one with the heatmap and a second one zooming on the current selection (done by selecting the top-left and bottom-right corner of the square you want to zoom in):

References

• 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.