Deep graph embedding for prioritizing synergistic anticancer drug combinations

Peiran Jiang; Shujun Huang; Zhenyuan Fu; Zexuan Sun; Ted M Lakowski; Pingzhao Hu

doi:10.1016/j.csbj.2020.02.006

Deep graph embedding for prioritizing synergistic anticancer drug combinations

Comput Struct Biotechnol J. 2020 Feb 15:18:427-438. doi: 10.1016/j.csbj.2020.02.006. eCollection 2020.

Authors

Peiran Jiang^{1

2}, Shujun Huang³, Zhenyuan Fu⁴, Zexuan Sun^{1

5}, Ted M Lakowski³, Pingzhao Hu^{1

6}

Affiliations

¹ Department of Biochemistry and Medical Genetics, University of Manitoba, Winnipeg, Manitoba R3E 0J9, Canada.
² Department of Bioinformatics & Systems Biology, Huazhong University of Science and Technology, Wuhan 430074, China.
³ College of Pharmacy, University of Manitoba, Winnipeg, Manitoba R3E 0T5, Canada.
⁴ Tongji Medical College, Huazhong University of Science and Technology, Wuhan 430030, China.
⁵ School of Mathematics and Statistic, Wuhan University, Wuhan 430072, China.
⁶ Research Institute in Oncology and Hematology, CancerCare Manitoba, Winnipeg R3E 0V9, Canada.

Abstract

Drug combinations are frequently used for the treatment of cancer patients in order to increase efficacy, decrease adverse side effects, or overcome drug resistance. Given the enormous number of drug combinations, it is cost- and time-consuming to screen all possible drug pairs experimentally. Currently, it has not been fully explored to integrate multiple networks to predict synergistic drug combinations using recently developed deep learning technologies. In this study, we proposed a Graph Convolutional Network (GCN) model to predict synergistic drug combinations in particular cancer cell lines. Specifically, the GCN method used a convolutional neural network model to do heterogeneous graph embedding, and thus solved a link prediction task. The graph in this study was a multimodal graph, which was constructed by integrating the drug-drug combination, drug-protein interaction, and protein-protein interaction networks. We found that the GCN model was able to correctly predict cell line-specific synergistic drug combinations from a large heterogonous network. The majority (30) of the 39 cell line-specific models show an area under the receiver operational characteristic curve (AUC) larger than 0.80, resulting in a mean AUC of 0.84. Moreover, we conducted an in-depth literature survey to investigate the top predicted drug combinations in specific cancer cell lines and found that many of them have been found to show synergistic antitumor activity against the same or other cancers in vitro or in vivo. Taken together, the results indicate that our study provides a promising way to better predict and optimize synergistic drug pairs in silico.

Keywords: ACC, accuracy; AUC, area under the curve; CNN, convolutional neural network; Cancer; Cell line; DDS, drug-drug synergy; DNN, deep neural network; DTI, drug-target interaction; ER, estrogen receptor; FPR, false positive rate; GBM, glioblastoma multiforme; GCN, graph convolutional network; Graph convolutional network; HTS, high throughput screening; Heterogenous network; PPI, protein–protein interaction; RF, random forest; ROC, receiver operating characteristic; SD, standard deviation; SVM, support vector machine; Synergistic drug combination; TNBC, triple negative breast cancer; TPR, true positive rate; XGBoost, extreme gradient boosting.