Limitations of samples for co-expression network analysis

Question

[f.dehghanian68]

Dear all,

I'm working on reconstruction and analysis of a co-expression network of a signal pathway in breast tumor. I have GEO data just for 10 patients with breast tumors with specified features. Are these number of sample suitable for network analysis? Is this network acceptable?

On the other hand I want to reconstruct a co-expression network for differentiation of ESCs to neural cells. GEO data is limited for this study. For example I have a dataset which contains expression profile of 4 days of differentiation , each contains 3 replicates. Is it a good dataset for my analysis for reconstruction of network?

Answer 1

[k.theofilatos]

You can construct co-expression networks from your gene expression experiments with various methods with InSyBio BioNets (a demo is available here and you can request for a free one month access at [e-mail]). Our experimental results indicated that at least 5 samples are needed to acquire meaningful results. For more details read the relevant white paper: http://insybio.com/pages/bionets