Table 2 Effect of different cutoff values for p-value on RATE results. A cutoff of p <0.01 gave the best results with MCC = 0.451
From: Experimental validation of the RATE tool for inferring HLA restrictions of T cell epitopes
Cutoffs | TP | FP | FN | TN | Total | Accuracy | Sensitivity | Specificity | Precision | False Positive Rate | Matthews Correlation Coefficient |
|---|---|---|---|---|---|---|---|---|---|---|---|
p <0.05 | 27 | 17 | 13 | 45 | 102 | 0.706 | 0.675 | 0.726 | 0.614 | 0.274 | 0.395 |
p <0.01 | 22 | 8 | 18 | 54 | 102 | 0.745 | 0.550 | 0.871 | 0.733 | 0.129 | 0.451 |
p <0.005 | 18 | 5 | 22 | 57 | 102 | 0.735 | 0.450 | 0.919 | 0.783 | 0.081 | 0.432 |