deeptools PCA not transposed

[laurie-tonon]

Hi,

I noticed that the PCA from the deeptools part of the pipeline is always strange, with the samples all aligned on one part of the graph, even though the first axis represents the majority of the variance.

For example:

It seemd to be related to this issue: deeptools/deepTools#1238, and the matrice output from multiBamSummary should be transposed, using the --transpose argument for the plotPCA tool.

Does this make sense to you?

Thanks