Table 1

Performance of classifiers trained on human versus mouse kinases in predicting AmiGO annotations. The performance measures accuracy, kappa coefficient, correlation coefficient, precision, and recall are reported for two of the HDTree classifiers. The first classifier is trained on 330 human kinases. The performance is based on 10-fold cross-validation. The second classifier is trained on the 330 human kinases and tested on 244 mouse kinases. The annotations for the mouse and human kinases were obtained from AmiGO.




Correlation Coefficient
Precision
Recall




Classifier
Accuracy
Kappa Coefficient
Ser/Thr
Tyr
Dual
Ser/Thr
Tyr
Dual
Ser/Thr
Tyr
Dual

Human
89.1
0.76
0.82
0.86
0.30
0.97
1.00
0.15
0.95
0.74
0.71
Mouse
15.1
-0.40
-0.40
-0.43
-0.01
0.17
0.11
0.25
0.41
0.07
0.01

Andorf et al. BMC Bioinformatics 2007 8:284   doi:10.1186/1471-2105-8-284