Skip to main content

Table 5 The performances of individual feature-based models on imbalanced Human dataset

From: A genetic algorithm-based weighted ensemble method for predicting transposon-derived piRNAs

Index Feature AUC ACC SN SP
F1 1-Spectrum Profile 0.748 0.739 0.398 0.854
F2 2-Spectrum Profile 0.841 0.808 0.416 0.940
F3 3-Spectrum Profile 0.850 0.814 0.321 0.982
F4 4-Spectrum Profile 0.844 0.811 0.284 0.989
F5 5-Spectrum Profile 0.836 0.813 0.305 0.986
F6 (3,1)-Mismatch Profile 0.867 0.824 0.427 0.959
F7 (4,1)-Mismatch Profile 0.856 0.814 0.328 0.979
F8 (5,1)-Mismatch Profile 0.851 0.810 0.277 0.991
F9 (3,1)-Subsequence Profile 0.850 0.808 0.443 0.932
F10 (4,1)-Subsequence Profile 0.864 0.822 0.473 0.940
F11 (5,1)-Subsequence Profile 0.871 0.829 0.492 0.944
F12 1-RevcKmer 0.745 0.746 0.005 0.997
F13 2-RevcKmer 0.803 0.778 0.411 0.902
F14 3-RevcKmer 0.823 0.800 0.265 0.981
F15 4-RevcKmer 0.823 0.803 0.241 0.993
F16 5-RevcKmer 0.818 0.806 0.255 0.992
F17 PCPseDNC 0.841 0.806 0.374 0.952
F18 PCPseTNC 0.857 0.813 0.337 0.975
F19 SCPseDNC 0.836 0.803 0.346 0.958
F20 SCPseTNC 0.842 0.808 0.312 0.977
F21 Sparse Profile 0.905 0.856 0.634 0.932
F22 PSSM 0.882 0.832 0.584 0.916
F23 LSSTE 0.688 0.766 0.175 0.966