Comparing coexpression results between two datasets

Question

[brendan.innes]

Apologies if this has been asked - I assumed it would have been, but I can't seem to find it.

I am trying to do a differential coexpression analysis, which is a well-covered topic. I've been looking at EBcoexpress (bioconductor package) as my preferred tool. The problem is, I'm trying to compare two different cancer types, so the classes for the differential analysis are represented by each of two datasets. In other words, I'm essentially comparing the coexpression network from dataset A to the coexpression network from dataset B. While the datasets are based on the same microarray platform, I'm still concerned that it's inappropriate to compare across microarrays like that.

Any ideas, input, and/or references that you're familiar with in which two completely independent datasets were compared to yield a differential coexpression measure would be super-appreciated.

Thanks!

Answer 1

[raunakms]

I think looking at tissue-specific expression would also be wise as if two cancer-types have different tissue composition.

Answer 2

[k.theofilatos]

You can construct co-expression networks from your gene expression experiments and compare them with various algorithms and methods (using degree centrality, clustering coefficient or random walk methods) with InSyBio BioNets (a demo is available here and you can request for a free one month access at [e-mail]). For more details read the relevant white paper: http://insybio.com/pages/bionets