how to build an interaction network in R

Question

[pinheirofabiano]

I'm investigating the role of my gene of interest (GOI) in cancer, using gene co-expression analysis (guilty by association). Performing gene co-expression analysis, I've detected my GOI associated with 163 other genes in a module.

I've already performed an enrichment analysis with the genes of this module and I know the potential pathways associated with this gene.

Now I would like to build an interaction network with the genes of this module to better understand how these genes are related and detect the major hubs.

Can you suggest me a package in R to do that? I would like to do it all in R, instead of using softwares as Ingenuity or similar approaches.

Answer 1

[rodolfo.peacewalker]

Hi!

I'm not familiar with gene co-expression analysis output. However, if you have a data frame consisting of edges and weights you could use some useful R packages such as igraph to perform network analysis (based on topological features). Take a look at this tutorial in order to meet the features to create an igraph object.

Best regards

[pinheirofabiano]

Hi, rodolfo.peacewalker

Can I use the genes identified in my gene co-expression analysis as nodes and the FPKM values from the RNA-Seq dataset as a parameter of node size to generate a network?

I believe it would be a complete network, since all nodes are probably interconnected, but I don't have how to obtain this information for sure..

many thanks, Fabiano

Answer 2

[Kevin Blighe]

There is also this: Network plot from expression data in R using igraph

Hub [and other] scores are mentioned in that tutorial. The function to use is this: https://igraph.org/r/doc/hub_score.html

Kevin