Table 3. Performance comparison among various types of sequence features and methods for the across-germline prediction in terms of Jackknife test accuracy.
Method | Sensitivity | Specificity | Accuracy (%) |
Random Forests (AAC) | 0.846 | 0.731 | 79.63 |
Random Forests (DPC) | 0.854 | 0.806 | 83.33 |
Random Forests (PPs) | 0.846 | 0.720 | 79.17 |
Random Forests (AAC+DPC) | 0.854 | 0.801 | 83.10 |
Random Forests (AAC+PPs) | 0.854 | 0.731 | 80.09 |
Random Forests (DPC+PPs) | 0.870 | 0.774 | 82.87 |
Random Forests (AAC+DPC+PPs) | 0.858 | 0.763 | 81.71 |