Table 3 Comparison of performance of classifiers based on AmiGO annotations and UniProt annotations. The performance measures accuracy, kappa coefficient, correlation coefficient, precision, and recall are reported for two of the HDTree classifiers. Both classifiers were trained on 330 human kinases and tested on 211 mouse kinases with RCA evidence codes in AmiGO. The first classifier was trained and tested with annotations provided by UniProt and the second classifier used annotations obtained from AmiGO.
From: Exploring inconsistencies in genome-wide protein function annotations: a machine learning approach
| Â | Â | Â | Correlation Coefficient | Precision | Recall | ||||||
|---|---|---|---|---|---|---|---|---|---|---|---|
Classifier | Accuracy | Kappa Coefficient | Ser/Thr | Tyr | Dual | Ser/Thr | Tyr | Dual | Ser/Thr | Tyr | Dual |
UniProt | 97.1 | 0.93 | 0.98 | 0.94 | 0.00 | 0.97 | 0.97 | 0.00 | 0.99 | 1.00 | 0.00 |
AmiGO | 4.2 | -0.37 | -0.64 | -0.85 | 0.00 | 0.06 | 0.00 | 0.00 | 0.14 | 0.00 | 0.00 |