Skip to main content

Table 2 Performance evaluation of the investigated features in identifying ubiquitylation sites based on five-fold cross-validation

From: UbiSite: incorporating two-layered machine learning method with substrate motifs to predict ubiquitin-conjugation site on lysines

Investigated features Sensitivity Specificity Accuracy MCC
20D binary coding (20D) 65.59 % 67.09 % 66.64 % 0.303
Amino Acid Composition (AAC) 64.34 % 65.44 % 65.11 % 0.275
Amino Acid Pair Composition (AAPC) 68.70 % 70.72 % 70.11 % 0.367
Position Weight Matrix (PWM) 68.08 % 67.99 % 68.01 % 0.334
Position-Specific Scoring Matrix (PSSM) 69.46 % 70.69 % 70.32 % 0.374
Solvent-Accessible Surface Area (SASA) 64.58 % 65.47 % 65.20 % 0.278
Secondary Structure (SS) 55.20 % 60.51 % 58.91 % 0.145