Table 1.
Performance comparison between the model with gene expression data alone and models identified using knowledge-based matrices
| Data type | Balanced accuracy | AUC |
|---|---|---|
| Gene expression |
0.6957 |
0.7103 |
| Pathway |
0.7451 |
0.7457 |
| GO |
0.6991 |
0.7275 |
| Pfam |
0.7046 |
0.7335 |
| Integration | 0.7882 | 0.8108 |
We compare here our results of evaluating gene-expression data alone, KEGG, GO, Pfam, and integration modeling. The integration model was developed by combining variables from KEGG pathway-based matrix, GO-based matrix, and Pfam-based matrix. Performances were measured based on the balanced accuracy and area under the ROC curve (AUC).