Table 2 Five-fold cross validation results on single SVM model trained with various features.
From: Characterization and identification of protein O-GlcNAcylation sites with substrate specificity
Training features | Number of positive data | Number of negative data | Sn | Sp | Acc | MCC |
|---|---|---|---|---|---|---|
20D Binary code | 375 | 375 | 0.66 | 0.69 | 0.69 | 0.21 |
Amino Acid Composition (AAC) | 375 | 375 | 0.64 | 0.65 | 0.65 | 0.17 |
Amino Acid Pair Composition (AAPC) | 375 | 375 | 0.66 | 0.67 | 0.67 | 0.20 |
Accessible Surface Area (ASA) | 375 | 375 | 0.57 | 0.59 | 0.59 | 0.10 |
Position Weight Matrix (PWM) | 375 | 375 | 0.62 | 0.63 | 0.63 | 0.14 |
Position-specific scoring matrix (PSSM) | 375 | 375 | 0.68 | 0.69 | 0.69 | 0.22 |