Fig 1.
An overview of MGAE-DC.
Table 1.
The synergy score regression performances on the O’Neil dataset, in terms of mean and standard deviation.
Table 2.
The classification performances of different methods on drug-drug synergy prediction on the O’Neil dataset using Loewe score, in terms of mean and standard deviation.
Fig 2.
The performances of different methods in each cell line.
(A) The RMSE and PCC of different methods in each cell line. The mean values of different folds are shown as solid lines. Error bars in terms of one standard deviation are shown as shaded areas. (B) The performances of different methods in the three representational cell lines. For each cell line, the scatter plot of MGAE-DC predicted synergy scores and the ground truth, the boxplots of RMSE and PCC of different methods, the RMSE of different methods in the synergistic and antagonistic drug combinations, respectively. The P-values are calculated by Wilcoxon signed-rank test. * represents P-value ≤ 0.05.
Table 3.
The results of the ablation study.
Fig 3.
Visualization of drug Zolinza cell line-specific embeddings in 2-dimensional space using t-SNE.
Table 4.
The performances of different predictors for MGAE-DC.
Table 5.
Summary of the drug combinations’ synergy data for different datasets.
Fig 4.
The schematics of different cross-validation strategies.