Possible to use regression for RNA-seq to predict targets of RNA-damaging drug?

Question

[dynev.aw]

First of all, I am new to the world of sequencing and bioinformatics and am mostly looking for guidance.

I have a drug that is cytotoxic to certain cancer cell lines and not toxic to 'healthy' cell lines. I want to use something like multiple regression on dataset where parameters are transcript expression values from RSEM and target is the cytotoxicity index, and do some sort of feature extraction after.

However, I am very new to RNA-seq and statistics, so I don't know if this idea is plausible at all and what software/methods to use if it is.

Thanks for attention.