Table 1.
Predictive accuracies using biomarker thresholds.
| Training Set | Test Set | Unit | TP | TN | FP | FN | Sensitivity | Specificity | PPV | NPV | Balanced accuracy |
|---|---|---|---|---|---|---|---|---|---|---|---|
| TG-Training | TG-Test | Chemical | 9 | 27 | 2 | 0 | 1.000 | 0.931 | 0.818 | 1.000 | 0.966 |
| TG-All | DM-All | Chemical | 33 | 23 | 20 | 1 | 0.971 | 0.535 | 0.623 | 0.958 | 0.753 |
| DM-All | TG-All | Chemical | 17 | 55 | 2 | 1 | 0.944 | 0.965 | 0.895 | 0.982 | 0.955 |
| DM-All | TG-4d | Chemical | 17 | 57 | 0 | 1 | 0.944 | 1.000 | 1.000 | 0.983 | 0.972 |
| DM-All | TG-8d | Chemical | 17 | 57 | 0 | 1 | 0.944 | 1.000 | 1.000 | 0.983 | 0.972 |
| DM-All | TG-15d | Chemical | 17 | 56 | 1 | 1 | 0.944 | 0.982 | 0.944 | 0.982 | 0.963 |
| DM-All | TG-29d | Chemical | 17 | 55 | 2 | 1 | 0.944 | 0.965 | 0.895 | 0.982 | 0.955 |
DM, DrugMatrix; TG, TG-GATES; TP, true positive; TN, true negative; FP, false positive; FN, false negative; PPV, positive predictive value; NPV, negative predictive value.