I am trying to analyze the differences in copy number alterations (CNA) between two classes of samples: cases and controls. Since GISTIC2.0 takes a group of samples and find focal CNAs, is it better to run GISTIC2.0 separately on my cases and controls, or should I run it on all samples combined?
My concern is that if i run GISTIC2.0 on all samples combined, some regions that are amplified/deleted only in the cases (or vice versa) might not show up in the results since GISTIC2.0 will only see amplifications/deletions in a smaller subset of the samples.
Any insight is appreciated! Thank you in advance.