The performance of SVM-based models developed on alternate dataset; models were developed using binary profile of terminal residues of peptides
| Techniques (Parameters) . | Training dataset . | Validation dataset . | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|
| Sen . | Spc . | Acc . | MCC . | AUROC . | Sen . | Spc . | Acc . | MCC . | AUROC . | |
| N5 (g = 0.5, c = 4) | 84.50 | 83.29 | 83.39 | 0.68 | 0.92 | 88.14 | 87.63 | 87.89 | 0.76 | 0.93 |
| N10 (g = 0.01, c = 3) | 85.95 | 83.14 | 84.58 | 0.69 | 0.92 | 88.48 | 86.63 | 87.60 | 0.75 | 0.94 |
| N15 (g = 0.01, c = 3) | 85.09 | 84.74 | 84.90 | 0.70 | 0.93 | 87.50 | 85.53 | 86.49 | 0.73 | 0.94 |
| C5 (g = 0.5, c = 1) | 82.30 | 75.84 | 79.06 | 0.58 | 0.88 | 83.51 | 86.08 | 84.79 | 0.70 | 0.90 |
| C10 (g = 0.01, c = 3) | 81.04 | 79.14 | 80.11 | 0.60 | 0.88 | 80.63 | 82.56 | 81.54 | 0.63 | 0.89 |
| C15 (g = 0.01, c = 3) | 83.02 | 84.90 | 84.03 | 0.68 | 0.90 | 81.25 | 88.82 | 85.14 | 0.70 | 0.91 |
| N5C5 (g = 0.01, c = 3) | 84.88 | 81.75 | 83.31 | 0.67 | 0.91 | 86.08 | 84.54 | 85.31 | 0.71 | 0.93 |
| N10C10 (g = 0.01, c = 4) | 87.99 | 87.43 | 87.72 | 0.75 | 0.94 | 85.34 | 91.28 | 88.15 | 0.77 | 0.95 |
| N15C15 (g = 0.01, c = 1) | 88.68 | 86.36 | 87.43 | 0.75 | 0.95 | 88.19 | 88.16 | 88.18 | 0.76 | 0.95 |
| Techniques (Parameters) . | Training dataset . | Validation dataset . | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|
| Sen . | Spc . | Acc . | MCC . | AUROC . | Sen . | Spc . | Acc . | MCC . | AUROC . | |
| N5 (g = 0.5, c = 4) | 84.50 | 83.29 | 83.39 | 0.68 | 0.92 | 88.14 | 87.63 | 87.89 | 0.76 | 0.93 |
| N10 (g = 0.01, c = 3) | 85.95 | 83.14 | 84.58 | 0.69 | 0.92 | 88.48 | 86.63 | 87.60 | 0.75 | 0.94 |
| N15 (g = 0.01, c = 3) | 85.09 | 84.74 | 84.90 | 0.70 | 0.93 | 87.50 | 85.53 | 86.49 | 0.73 | 0.94 |
| C5 (g = 0.5, c = 1) | 82.30 | 75.84 | 79.06 | 0.58 | 0.88 | 83.51 | 86.08 | 84.79 | 0.70 | 0.90 |
| C10 (g = 0.01, c = 3) | 81.04 | 79.14 | 80.11 | 0.60 | 0.88 | 80.63 | 82.56 | 81.54 | 0.63 | 0.89 |
| C15 (g = 0.01, c = 3) | 83.02 | 84.90 | 84.03 | 0.68 | 0.90 | 81.25 | 88.82 | 85.14 | 0.70 | 0.91 |
| N5C5 (g = 0.01, c = 3) | 84.88 | 81.75 | 83.31 | 0.67 | 0.91 | 86.08 | 84.54 | 85.31 | 0.71 | 0.93 |
| N10C10 (g = 0.01, c = 4) | 87.99 | 87.43 | 87.72 | 0.75 | 0.94 | 85.34 | 91.28 | 88.15 | 0.77 | 0.95 |
| N15C15 (g = 0.01, c = 1) | 88.68 | 86.36 | 87.43 | 0.75 | 0.95 | 88.19 | 88.16 | 88.18 | 0.76 | 0.95 |
Sen: sensitivity, Spc: specificity, Acc: accuracy, MCC: Matthews correlation coefficient, AUROC: area under the receiver operating characteristic curve, N5/N10/N15: First 5/10/15 elements from N-terminal, C5/C10/C15: First 5/10/15 elements from C-terminal, N5C5/N10C10/N15C15: First 5/10/15 elements from N-terminal as well as from C-terminal joined together.
The performance of SVM-based models developed on alternate dataset; models were developed using binary profile of terminal residues of peptides
| Techniques (Parameters) . | Training dataset . | Validation dataset . | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|
| Sen . | Spc . | Acc . | MCC . | AUROC . | Sen . | Spc . | Acc . | MCC . | AUROC . | |
| N5 (g = 0.5, c = 4) | 84.50 | 83.29 | 83.39 | 0.68 | 0.92 | 88.14 | 87.63 | 87.89 | 0.76 | 0.93 |
| N10 (g = 0.01, c = 3) | 85.95 | 83.14 | 84.58 | 0.69 | 0.92 | 88.48 | 86.63 | 87.60 | 0.75 | 0.94 |
| N15 (g = 0.01, c = 3) | 85.09 | 84.74 | 84.90 | 0.70 | 0.93 | 87.50 | 85.53 | 86.49 | 0.73 | 0.94 |
| C5 (g = 0.5, c = 1) | 82.30 | 75.84 | 79.06 | 0.58 | 0.88 | 83.51 | 86.08 | 84.79 | 0.70 | 0.90 |
| C10 (g = 0.01, c = 3) | 81.04 | 79.14 | 80.11 | 0.60 | 0.88 | 80.63 | 82.56 | 81.54 | 0.63 | 0.89 |
| C15 (g = 0.01, c = 3) | 83.02 | 84.90 | 84.03 | 0.68 | 0.90 | 81.25 | 88.82 | 85.14 | 0.70 | 0.91 |
| N5C5 (g = 0.01, c = 3) | 84.88 | 81.75 | 83.31 | 0.67 | 0.91 | 86.08 | 84.54 | 85.31 | 0.71 | 0.93 |
| N10C10 (g = 0.01, c = 4) | 87.99 | 87.43 | 87.72 | 0.75 | 0.94 | 85.34 | 91.28 | 88.15 | 0.77 | 0.95 |
| N15C15 (g = 0.01, c = 1) | 88.68 | 86.36 | 87.43 | 0.75 | 0.95 | 88.19 | 88.16 | 88.18 | 0.76 | 0.95 |
| Techniques (Parameters) . | Training dataset . | Validation dataset . | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|
| Sen . | Spc . | Acc . | MCC . | AUROC . | Sen . | Spc . | Acc . | MCC . | AUROC . | |
| N5 (g = 0.5, c = 4) | 84.50 | 83.29 | 83.39 | 0.68 | 0.92 | 88.14 | 87.63 | 87.89 | 0.76 | 0.93 |
| N10 (g = 0.01, c = 3) | 85.95 | 83.14 | 84.58 | 0.69 | 0.92 | 88.48 | 86.63 | 87.60 | 0.75 | 0.94 |
| N15 (g = 0.01, c = 3) | 85.09 | 84.74 | 84.90 | 0.70 | 0.93 | 87.50 | 85.53 | 86.49 | 0.73 | 0.94 |
| C5 (g = 0.5, c = 1) | 82.30 | 75.84 | 79.06 | 0.58 | 0.88 | 83.51 | 86.08 | 84.79 | 0.70 | 0.90 |
| C10 (g = 0.01, c = 3) | 81.04 | 79.14 | 80.11 | 0.60 | 0.88 | 80.63 | 82.56 | 81.54 | 0.63 | 0.89 |
| C15 (g = 0.01, c = 3) | 83.02 | 84.90 | 84.03 | 0.68 | 0.90 | 81.25 | 88.82 | 85.14 | 0.70 | 0.91 |
| N5C5 (g = 0.01, c = 3) | 84.88 | 81.75 | 83.31 | 0.67 | 0.91 | 86.08 | 84.54 | 85.31 | 0.71 | 0.93 |
| N10C10 (g = 0.01, c = 4) | 87.99 | 87.43 | 87.72 | 0.75 | 0.94 | 85.34 | 91.28 | 88.15 | 0.77 | 0.95 |
| N15C15 (g = 0.01, c = 1) | 88.68 | 86.36 | 87.43 | 0.75 | 0.95 | 88.19 | 88.16 | 88.18 | 0.76 | 0.95 |
Sen: sensitivity, Spc: specificity, Acc: accuracy, MCC: Matthews correlation coefficient, AUROC: area under the receiver operating characteristic curve, N5/N10/N15: First 5/10/15 elements from N-terminal, C5/C10/C15: First 5/10/15 elements from C-terminal, N5C5/N10C10/N15C15: First 5/10/15 elements from N-terminal as well as from C-terminal joined together.
This PDF is available to Subscribers Only
View Article Abstract & Purchase OptionsFor full access to this pdf, sign in to an existing account, or purchase an annual subscription.
