Hello,
I applied iClusterPlus to my data, in order to find the most significant driver cpgs and genes. I got two heatmaps: one for expression and one for methylation, and I got also the ids of the significant genes and cpgs, clearly distributed in 4 clusters.
How could I establish if there is actuallly a relationship between these genes and these cpgs modifications (differential expression, methylation)? I tried to check if those cpgs are indeed located in enhancers that have my significant genes as targets, or if those cpgs are close to my significant genes.
All in all, I got to the end of iClusterPlus manual and it is not clear to me which relationship lays between the genes and the cpgs. Maybe someone has used it and knows what the next step is?
Thank you very much.
Have you tried plotting expression as a function of methylation in those CpGs? That would seem to be the simplest way to look at this.
That seems a very interesting idea! But i am not sure how I could do this. I have beta values for each CpG and for each sample (19 tumor, 19 peri-tumor). And I have the expression of each gene for the same samples. I would have to plot each pair cpg-gene (all the combinations) of the significant genes and significant cpgs obtained for each of the 4 clusters right?
You'd presumably just color the results by cluster or something like that.