Research Article
Accurate Identification of Antioxidant Proteins Based on a Combination of Machine Learning Techniques and Hidden Markov Model Profiles
Table 2
Performance comparisons on the D1 dataset.
| Methods | Sen (%) | Spe (%) | Acc (%) | MCC | AUC |
| Naïve Bayes [14] | 72.04 | 66.05 | 66.88 | — | 0.855 | AodPred [15] | 75.09 | 74.48 | 74.79 | — | — | iANOP-Enble [48] | 72.78 | 90.75 | 88.25 | 0.5725 | 0.935 | SeqSVM [19] | — | — | 89.46 | — | — | IDAod3 [20] | 81.27 | 99.59 | 97.05 | 0.7409 | — | AOPs-SVM [18] | 68.0 | 98.5 | 94.2 | 0.741 | 0.832 | Vote9 [9] | 66.4 | 98.6 | 94.1 | 0.740 | — | Random forest [23] | 81.5 | 85.1 | 84.6 | — | — | Hybrid feature [24] | 66.50 | 96.30 | 83.91 | — | — | AOP-HMM | 98.23 | 97.78 | 98.01 | 0.9601 | 0.992 |
|
|
Note: results of iANOP-Enble and random forest were obtained by performing the 10-fold CV.
|