Table 2

The performance of SVM models developed using amino acid and PSSM matrix composition.
Thr
 Split Amino acids composition
 PSSM Matrix composition


Sn*
 Sp
 Acc
 MCC
 Sn
 Sp
 Acc
 MCC
-1.0
 93.25
 43.55
 68.60
 0.42
 97.22
 44.84
 71.03
 0.49
-0.9
 93.25
 50.00
 71.80
 0.48
 95.63
 58.73
 77.18
 0.58
-0.8
 92.46
 56.05
 74.40
 0.52
 95.24
 65.08
 80.16
 0.63
-0.7
 91.27
 59.68
 75.60
 0.54
 94.05
 70.24
 82.14
 0.66
-0.6
 90.48
 62.10
 76.40
 0.55
 94.05
 75.00
 84.52
 0.70
-0.5
 87.70
 64.11
 76.00
 0.53
 92.86
 84.13
 88.49
 0.77
-0.4
 85.71
 70.16
 78.00
 0.57
 91.27
 90.08
 90.67
 0.81
-0.3
 84.92
 73.39
 79.20
 0.59
 90.08
 91.27
 90.67
 0.81
-0.2
 83.73
 78.63
 81.20
 0.62
 90.08
 92.86
 91.47
 0.83
-0.1
 83.33
 82.26
 82.80
 0.66
 89.29
 94.05
 91.67
 0.83
0.0
 81.75
 87.50
 84.60
 0.69
 89.29
 94.84
 92.06
 0.84
0.1
 79.37
 93.55
 86.40
 0.74
 88.89
 95.24
 92.06
 0.84
0.2
 77.38
 94.76
 86.00
 0.73
 88.49
 96.83
 92.66
 0.86
0.3
 76.19
 96.37
 86.20
 0.74
 87.30
 97.22
 92.26
 0.85
0.4
 73.81
 97.98
 85.80
 0.74
 85.71
 97.22
 91.47
 0.83
0.5
 71.43
 97.98
 84.60
 0.72
 85.32
 98.41
 91.87
 0.84
0.6
 69.84
 97.98
 83.80
 0.71
 84.92
 98.81
 91.87
 0.85
0.7
 68.25
 97.98
 83.00
 0.69
 83.73
 98.81
 91.27
 0.83
0.8
 65.87
 97.98
 81.80
 0.67
 82.94
 99.21
 91.07
 0.83
0.9
 63.10
 98.39
 80.60
 0.66
 80.16
 99.21
 89.68
 0.81
1.0
 58.73
 99.60
 79.00
 0.64
 73.41
 100.00
 86.71
 0.76

*Sn: Sensitivity; Sp: Specificity; Acc: Accuracy; MCC: Mathews Correlation Coefficient

Verma et al. BMC Bioinformatics 2008 9:201   doi:10.1186/1471-2105-9-201