DrugSim2DR: systematic prediction of drug functional similarities in the context of specific disease for drug repurposing

Abstract

Background: Traditional approaches to drug development are costly and involve high risks. The drug repurposing approach can be a valuable alternative to traditional approaches and has therefore received considerable attention in recent years.

Findings: Herein, we develop a previously undescribed computational approach, called DrugSim2DR, which uses a network diffusion algorithm to identify candidate anticancer drugs based on a drug functional similarity network. The innovation of the approach lies in the drug-drug functional similarity network constructed in a manner that implicitly links drugs through their common biological functions in the context of a specific disease state, as the similarity relationships based on general states (e.g., network proximity or Jaccard index of drug targets) ignore disease-specific molecular characteristics. The drug functional similarity network may provide a reference for prediction of drug combinations. We describe and validate the DrugSim2DR approach through analysis of data on breast cancer and lung cancer. DrugSim2DR identified some US Food and Drug Administration-approved anticancer drugs, as well as some candidate drugs validated by previous studies in the literature. Moreover, DrugSim2DR showed excellent predictive performance, as evidenced by receiver operating characteristic analysis and multiapproach comparisons in various cancer datasets.

Conclusions: DrugSim2DR could accurately assess drug-drug functional similarity within a specific disease context and may more effectively prioritize disease candidate drugs. To increase the usability of our approach, we have developed an R-based software package, DrugSim2DR, which is freely available on CRAN (https://CRAN.R-project.org/package=DrugSim2DR).

Keywords: computational drug repurposing; drug–drug similarity; network analysis; specific disease state.

Figure 1:

Schematic overview of the DrugSim2DR method.

Figure 2:

Assessment of drug–drug functional similarity in the context of breast cancer. (A, B) Comparations of chemical structure similarity and semantic-based similarity across 4 groups (Q1–Q4) were categorized according to the quartiles of the functional similarity scores. Statistical significance between boxplots is calculated by Wilcoxon rank-sum tests (^****P < 0.0001). (C–E) Bipartite networks of drugs and their shared GO terms for ouabain/tegoprazan, ouabain/bisacodyl, and bisacodyl/tegoprazan pairs. The red nodes in the network represent drugs and the blue ones represent GO terms.

Figure 3:

Heatmap of gene expression levels of drugs’ targets between breast cancer and normal samples.

Figure 4:

Performance of the DrugSim2DR approach. (A) ROC curves for drugs identified by DrugSim2DR in 5 different cancer types. The AUROC values for drugs in each cancer type are calculated and displayed respectively. (B) Comparison of DrugSim2DR with 3 other approaches. We apply DrugSim2DR to 3 cancer types to compare the performance with the CMap, SubtypeDrug, and DvD. AUROC values are used to compare their performance.

Figure 5:

Robustness and reproducibility analysis of DrugSim2DR on breast cancer dataset. (A) Radar chart showing the overlapped number of top 50 drugs identified based on the restart probability r values set from 0.1 to 0.8 compared with that of r = 0.9. (B) Boxplots showing the AUROC values of predicted drugs for the different data removal. The red line indicates the AUROC value of the original data. (C) Venn diagram showing the overlapped number of the top 50 drugs identified in the GSE53752, GSE42568, and GSE21422 datasets. Correlation analysis of DrugSim2DR’s predictions for breast cancer across different datasets: (D) GSE53752 and GSE42568, (E) GSE53752 and GSE21422, and (F) GSE42568 and GSE21422.