Skip to main content
. 2014 May 10;9:2225–2239. doi: 10.2147/IJN.S57526

Table 3.

Performance of individual classifiers with different resampled data sizes in terms of accuracy (%)

Kinase SVM
MLP
K-NN
B=10 B=15 B=20 B=10 B=15 B=20 B=10 B=15 B=20
PKA 66.25 65.43 63.44 55.67 60.06 56.85 59.25 59.72 56.94
PKB 71.69 78.61 66.98 55.42 69.26 59.09 67.61 68.57 63.8
PKC 68.85 61.72 60.76 64.84 57.88 54.7 61.33 61.09 55.36
MAPK 61.24 67.67 62.02 57.52 58.72 58.67 57.41 60.91 61.45
GSK-3 67 73.16 66.5 59.5 56.41 58.12 68.75 66.25 71.25
CDK1 83.8 65.53 65.9 75.5 60.26 58.22 77.5 61 57
CK2 67.35 64.9 67.42 57.36 59.69 56.97 58 64.41 60.49
CAM KII 78.2 71.2 73.6 61.98 61.58 56.788 74.25 60.39 68.31
SRC 66.44 65.48 68.55 58.83 58.18 55.3 63.88 75 60

Note: Physicochemical properties of amino acids are used for sequence encoding.

Abbreviations: CAM KII, calmodulin-dependent protein kinase II; CDK1, cyclin-dependent kinase 1; CK2, casein kinase 2; GSK-3, glycogen synthase kinase 3; K-NN, K-nearest neighbor; MAPK, mitogen-activated protein kinase; MLP, multilayer perceptron; PKA, protein kinase A; PKB, protein kinase B; PKC, protein kinase C; SRC, tyrosin kinase SRC; SVM, support vector machine.