Dear Bioconductor users,

I am working with TCGA RNA-seq data. I performed WGCNA analysis using vst transformed data and I found 13 network modules. To explore whether the most significant representatives (module membership (kME)>=0.7) in each module share a common biological function, I performed network crosstalk based KEGG pathway enrichment analysis using BinoX algorithm implemented in the newly developed pathwAX web server to analyze the enrichment of network links between each query gene list and KEGG pathways. Is it true to perform network crosstalk based pathway analysis of transcriptional nework modules in functional association networks? 

 Now I want to construct the functional association network of potential  interactions (direct and indirect) between genes included in each module. In my case, would you recommend me to use the FUNCOUP Network or something else such as STRING database? Should I use expansion step and which is the most appropriate confidence level? Should I consider the significant representatives of WGCNA modules (kME>=0.7)  as "bad" genes if they have not any link?

I would appreciate hearing your opinion!!

Thank you very much for your time in advance!!

Sincerely,

Panagiotis Mokos