clustering gene expression data

Question

I have a question about clustering. I' m managing gene expression microarray data and I would like to cluster them in classes. I searched around to find the best clustering algorithm for my data, but since as in all gene expression microarray experiments the number of genes exceed greatly the number of samples being analyzed, the majority of this algorithms assume independency of the variables (so the genes). But in reality we all know that the activity of one gene is not independent from the activity of another gene. My question is: is there an algorithm or a technique able to take into account with simplifications such a dependencies between genes?

Thanks a lot!

Best

e.

Answer 1

This is a biclustering problem. Here's a paper that may be relevant for you: Biclustering of Expression Data.

Answer 2

There is a huge amount of literature on biclustering of gene expression data. Obviously various stuff has been tried. It's just that the simpler models (assuming independency etc.) are usually more efficient, i.e. run in reasonable time instead of days if you have a lot of data.

Either way, please look up the latest literature yourself. Any answer we give here will be outdated in half a year by newer publications on this matter.