Hello R gurus.
I have been performing hierarchical clustering of gene expression data. I use
the heatmap() function to perform the clustering and overlay classifications by
using the ColSideColors parameter.
In many cases I have various metadata associated with my samples, which leaves
me with many different sets of sample classes which can be shown with
ColSideColors. So my question is: Is there an R package available that allows
one to calculate which of the pre-defined sample classifications correlates best
with the hierarchical clusters? Having a tool that finds the most relevant
classification would save me the effort of visualise them all on a heatmap and
work out by eye whether there is a correlation or not. I don't doubt there
is a function that does this, but I haven't been able to find it - probably
due to me not using the correct terminology in my searches.
Many thanks for your help,
Pall
--------------------------------------------------------------------------
AstraZeneca UK Limited is a company incorporated in Engl...{{dropped:21}}