Hi all,
I have done a GSVA on my RNA-Seq data. For some of these samples we also have drug screening done. I want to see if the effective drugs can be matched to certain pathway alteration. I am just wondering how I can identify altered pathway in a single sample.
I understand that we can use linear regression model like limma to do group comparison. But a lot of the drugs are only effective to one sample and can't be grouped. I would appreciate if someone could point me to the right group of test to do this.
My limited statistical background tells me GSVA enrichment score is gaussian across sample but not zero centered. Would it make sense to just convert them all to z-score and pick those on the two ends if I just want to look at a single sample?
We use drug sensitivity score to catch sensitive drugs from the assay. I assume we can perform some correlation test between GSVA and sensitivity score, but what correlation test would be appropriate? I am not sure if that can be done if the drugs that are interesting usually work on one sample only and there isn't any "correlation" that can be seen with correlation test.
Thanks a lot!
