Skip to main content

Table 3 Performance of models based on both positive dataset and negative 1 dataset using SMO as classifier

From: Incorporating support vector machine with sequential minimal optimization to identify anticancer peptides

Features Training Testing
Sensitivity Specificity Accuracy MCC Sensitivity Specificity Accuracy MCC
AAC 0.756 0.888 0.822 0.587 0.760 0.840 0.800 0.556
N5C5 0.700 0.808 0.754 0.511 0.660 0.820 0.740 0.486
k-space = 0 0.790 0.838 0.814 0.629 0.830 0.860 0.845 0.690
k-space = 1 0.834 0.868 0.851 0.702 0.830 0.860 0.845 0.690
k-space = 2 0.812 0.877 0.844 0.690 0.770 0.800 0.785 0.570
AAC + k-space = 0 0.840 0.793 0.816 0.634 0.850 0.860 0.855 0.710
AAC + N5C5 0.728 0.873 0.800 0.607 0.720 0.860 0.790 0.586
N5C5 + k-space = 0 0.784 0.834 0.809 0.618 0.820 0.840 0.830 0.660
AAC + N5C5 + k-space 0 0.793 0.849 0.821 0.642 0.830 0.860 0.845 0.690
PSSM 0.844 0.862 0.853 0.706 0.850 0.800 0.825 0.651