Hi all! I'm analyzing patient-derived RNA-seq data from TCGA.

I performed a differential expression analysis to identify which cancer types show significant expression differences (I have a specific gene), compared to healthy normal tissue samples from the given cancer project. I compared not only matched controls by barcode, but everything I found. Now, I would like to identify the number of patients and their percentage in some cancer types (cancer samples only), with the "overexpression" of this specified gene. I am little bit confused about this task, how should I do it precisely?