Deep learning prediction of chemical-induced dose-dependent and context-specific multiplex phenotype responses and its application to personalized alzheimer’s disease drug repurposing
Fig 3
Comparison of drug perturbed gene expression profile prediction performance of DeepCE and MultiDCP.
(A) The average Pearson correlation and Spearman correlation of predicted gene expression profile and ground truth are used as the evaluation metrics. (B) Comparison of drug perturbed gene expression profile prediction precision of top upregulated genes (Pre@K (up)), and top downregulated genes (Pre@K (down)) of DeepCE and MultiDCP. We selected the top 10 and 100 upregulated and downregulated genes and evaluated their prediction precisions. **** indicates that the p-value of the two-tailed t-test is less than 1.0x10-4.