Research Article
CirBiTree: Citrullination Site Inference Based on a Fuzzy Neural Network and Flexible Neural Tree
Table 5
The performances of different features in length 19.
| Features | Sn (%) | Sp (%) | F1 | MCC |
| Binary encoding | 56.36 | 75.56 | 0.6234 | 0.3252 | AA composition | 64.77 | 62.79 | 0.6413 | 0.2756 | Grouping AA composition | 71.75 | 71.85 | 0.7178 | 0.4359 | Physicochemical properties | 75.36 | 73.73 | 0.7475 | 0.4910 | KNN features | 74.87 | 65.63 | 0.7157 | 0.4068 | Secondary tendency structure | 69.85 | 77.38 | 0.7258 | 0.4736 | PSSM | 71.17 | 79.29 | 0.7418 | 0.5062 | BPB | 72.68 | 78.45 | 0.7484 | 0.5121 | Proposed algorithm | 81.01 | 80.09 | 0.8064 | 0.6111 |
|
|