Skip to main content

Table 3 Performance of based models using individual feature and their combination, evaluated by LOOCV

From: Prediction of conformational B-cell epitopes from 3D structures by random forests with a distance-based feature

  Bound dataset Unbound dataset
Feature AUC F ACC SN SP AUC F ACC SN SP
Conservation 0.554 0.186 0.634 0.418 0.660 0.558 0.188 0.604 0.445 0.634
Composition 0.563 0.163 0.747 0.275 0.801 0.566 0.147 0.708 0.244 0.774
Secondary structure 0.510 0.161 0.627 0.297 0.733 0.531 0.167 0.642 0.372 0.690
Accessible area 0.570 0.192 0.593 0.470 0.605 0.618 0.241 0.562 0.616 0.548
ARD 0.589 0.206 0.620 0.487 0.633 0.627 0.243 0.585 0.578 0.588
Combination 0.633 0.227 0.672 0.490 0.692 0.654 0.256 0.608 0.601 0.606