0

I'm trying to compute the mutual information conditioned by a third variable between expression and methylation data for each gene in a dataset.

My data consist in a list with two matrices whose content is, for the first element, the RPKM expression values of 10000 genes measured in 35 patients that are suffering lung adenocarcinoma. The second element of the list, is another matrix composed by the beta methylation values of the same 10000 genes in 35 patients too. Both matrices have the genes in columns and pacients in rows.

So, I would like to compute the mutual information between methylation and expression values for all pairwise genes in both matrices. But, at the same time, this two variables are conditioned by a methylation threshold whose values oscillates between 0 and 1. I have seen the condinformation() function, but I don't be able to apply this function to every pairwise methylation and expression information for every gene.

Thank you so much!

Dani
  • 153
  • 8
  • First, can you define "mutual information"? Second, it sounds like you're looking for `outer`. – Hong Ooi Jun 19 '15 at 10:33
  • The mutual information is a measure of the mutual dependence between two random variables. You can see more information in https://en.m.wikipedia.org/wiki/Mutual_information I don't know what you want to say with "outer"... – Dani Jun 19 '15 at 11:01
  • The `entropy` package might be helpful. However, unless you post a good minimal reproducible example, it will be hard to find help here http://stackoverflow.com/questions/5963269/how-to-make-a-great-r-reproducible-example – rmuc8 Jun 19 '15 at 11:21

0 Answers0