Prediction of prognostic signatures in triple-negative breast cancer based on the differential expression analysis via NanoString nCounter immune panel

Table 4 Performance comparison with our model and Elastic Net model

^aAUC: Receiver Operating Characteristic Area Under Curve
^bPPV: Predictive Positive Value (TP / TP + FP)
^cOur model pCR: Random Forest analysis using pCR DEG.
^dOur model RELAPSE: Random Forest analysis using RELAPSE DEG.
^eEN model pCR: Random Forest analysis using EN pCR genes (alpha value < 0.95)
^fEN model RELAPSE: Random Forest analysis using EN RELAPSE genes (alpha value < 0.2)

ISSN: 1471-2407