How To Do Find Clusters Most Correlated With Phenotype?

Question

[Diwan]

Hi all,

Maybe it is a simple question, but any help is appreciated.

I have a microarray dataset of 5 treatments/5 controls(in duplicate). The only experimental value for the treatment is, it has a score 2 for the disease whereas controls have score 0. I have used 5 treatments vs. 8K genes to cluster the genes. I would like to find out which cluster of genes are most correlated with the phenotype. Is there a way to do it?

Edit: The experiment is: rats are treated with 5 chemicals(duplicates) and they have corresponding controls without chemical treatment. These chemicals produce liver inflammation. The phenotype is measured as histopathology scores (discrete values) between 0 to 3. All treatments has a score of 2. After clustering I would like to select clusters that are most relevant to phenotype.

Thanks,

Diwan

[Eric Normandeau]

It would be nice if you could provide more information about your design, specifically about the phenotypes and what kind of values (discreet or continuous) it can take. Please add the clarifications directly to the question (rather than as a reply to this comment).

Answer 1

[Eric Normandeau]

Depending ow what you are trying to do, the WGCNA package in R may be of help to you.

Description from the site:

The WGCNA R software package is a comprehensive collection of R functions for performing various aspects of weighted correlation network analysis. The package includes functions for network construction, module detection, gene selection, calculations of topological properties, data simulation, visualization, and interfacing with external software. Along with the R package we also present R software tutorials. While the methods development was motivated by gene expression data, the underlying data mining approach can be applied to a variety of different settings.

[Diwan]

Thanks Eric. Will check that. -Diw