Hi! I am working with data from the TCGA project, and noticed that a lot of genes have mutations in the 5 and 3 unstralated regions and would like to keep the genes with these mutations that might be affecting gene expression and discard the ones that are unlikely. Please notice that I am not interested in stablishing causality with this, just to separate genes whose mutations are highly likely to be affecting their expression. So, given that these mutations are most probably affecting expression at the translation level, I thought that it could be indicative to calculate a ratio between z-scores of the mRNA levels and the protein level, but I am not pretty sure that a comparison like that would be valid to make, so would like to know if this make any sense to you, or if you know a better method to do that. Thanks in advance.
