Codes for analysis of TNBC nanostring immune panel data using edgeR and randomForests
This study describes an approach for biomarker discovery, which predicts relapse and pCR in TNBC, by a learning prediction model, using a random forest with features selected from differential gene expression for the NanoString nCounter immune panel. To overcome a small sample size limitation and build prediction models, random forest model is constructed on the differentially expressed genes (DEGs) as selected features using edgeR.