Table 2. External Test Resultsa.
| n | AUC | BAC | ACC | sensitivity | specificity | F-measure | precision | recall | MCC | |
|---|---|---|---|---|---|---|---|---|---|---|
| MD-based method | 309 | 0.883 | 0.817 | 0.825 | 0.772 | 0.863 | 0.784 | 0.797 | 0.772 | 0.638 |
| DeepSnap-DL | 309 | 0.905 | 0.833 | 0.832 | 0.843 | 0.824 | 0.805 | 0.770 | 0.843 | 0.659 |
| ensemble model | 309 | 0.943 | 0.868 | 0.874 | 0.835 | 0.901 | 0.845 | 0.855 | 0.835 | 0.739 |
| consensus model | 214 | 0.958 | 0.959 | 0.953 | 0.963 | 0.948 | 0.943 | 0.953 | 0.915 |
MD-based method, molecular descriptor-based method; AUC, area under the curve of the receiver operating characteristic curve; BAC, balanced accuracy; ACC, accuracy; MCC, Matthews correlation coefficient.