Research Article
LipoFNT: Lipoylation Sites Identification with Flexible Neural Tree
Table 1
The Performances of Different Features.
| Features | Sn(%) | Sp(%) | Acc(%) | F1 | MCC |
| Binary Encoding | 56.36 | 75.80 | 66.08 | 0.6243 | 0.3279 | AA Composition | 64.84 | 62.79 | 63.82 | 0.6418 | 0.2764 | Grouping AA Composition | 71.78 | 72.04 | 71.91 | 0.7187 | 0.4382 | Physicochemical Properties | 75.53 | 73.93 | 74.73 | 0.7493 | 0.4947 | KNN Features | 74.94 | 65.85 | 70.40 | 0.7168 | 0.4096 | Secondary Tendency Structure | 69.96 | 77.40 | 73.68 | 0.7266 | 0.4749 | PSSM | 71.20 | 79.39 | 75.30 | 0.7424 | 0.5076 | BPB | 72.81 | 78.51 | 75.66 | 0.7495 | 0.5140 | Bi-gram | 75.17 | 76.81 | 75.99 | 0.7579 | 0.5199 | Tri-gram | 77.28 | 78.27 | 77.78 | 0.7766 | 0.5555 | Proposed Algorithm | 81.07 | 80.29 | 80.68 | 0.8076 | 0.6136 |
|
|