Table 1 Performance comparison
From: DualGCN: a dual graph convolutional network model to predict cancer drug response
Method | Pearson’s correlation | Spearman’s correlation | RMSE |
|---|---|---|---|
SVM | 0.336 ± 0.078 | 0.230 ± 0.071 | 3.115 ± 0.053 |
Random Forest | 0.864 ± 0.001 | 0.839 ± 0.003 | 1.441 ± 0.008 |
Lasso | 0.893 ± 0.002 | 0.873 ± 0.002 | 1.284 ± 0.007 |
Ridge | 0.895 ± 0.002 | 0.875 ± 0.002 | 1.268 ± 0.007 |
DeepCDR (-) | 0.900 ± 0.004 | 0.877 ± 0.004 | 1.265 ± 0.020 |
CDRscan | 0.911 ± 0.002 | 0.894 ± 0.002 | 1.173 ± 0.011 |
DualGCN | 0.925 ± 0.001 | 0.907 ± 0.002 | 1.079 ± 0.007 |
DeepCDR | 0.928 ± 0.001 | 0.910 ± 0.001 | 1.066 ± 0.004 |