Gold Standards or Evaluation Criteria for Gene Regulatory Network of Cancer Cell Types (Breast Cancer, ...)

Question

[ilikebing2000]

Hi everyone!

I've inferred 6 GRNs (Gene Regulatory Networks) using CLR (Context Likelihood Relatedness) method for breast cancer and 5 other cancer types using RNA-seq gene expression data.

Now how I can evaluate the inferred GRNs? I've searched but I don't find any gold standard or evaluation criterion?

Thanks.

[Bioinformatist Newbie]

You can use experimental ChIP-seq data for the TFs expressed in that particular cell type. For example if you have inferred a network for K562 or HepG2 cancer cell types then in the ENCODE, you can find ChIP-seq data for around 200 different TFs in K562 cell type. See what interactions are being suggested by that ChIP-seq experiment and what is their enrichment in your networks.