Abstract
Drug combination therapies are a promising strategy to overcome drug resistance and improve the efficacy of monotherapy in cancer, and it has been shown to lead to a decrease in dose-related toxicities. Except the synergistic reaction between drugs, some antagonistic drug–drug interactions (DDIs) exist, which is the main cause of adverse drug events. Precisely predicting the type of DDI is important for both drug development and more effective drug combination therapy applications. Recently, numerous text mining– and machine learning–based methods have been developed for predicting DDIs. All these methods implicitly utilize the feature of drugs from diverse drug-related properties. However, how to integrate these features more efficiently and improve the accuracy of classification is still a challenge. In this paper, we proposed a novel method (called NMDADNN) to predict the DDI types by integrating five drug-related heterogeneous information sources to extract the unified drug mapping features. NMDADNN first constructs the similarity networks by using the Jaccard coefficient and then implements random walk with restart algorithm and positive pointwise mutual information for extracting the topological similarities. After that, five network-based similarities are unified by using a multimodel deep autoencoder. Finally, NMDADNN implements the deep neural network (DNN) on the unified drug feature to infer the types of DDIs. In comparison with other recent state-of-the-art DNN-based methods, NMDADNN achieves the best results in terms of accuracy, area under the precision-recall curve, area under the ROC curve, F1 score, precision and recall. In addition, many of the promising types of drug–drug pairs predicted by NMDADNN are also confirmed by using the interactions checker tool. These results demonstrate the effectiveness of our NMDADNN method, indicating that NMDADNN has the great potential for predicting DDI types.
Highlights
Combined drug therapies are becoming the prevalent approach for complex disease in recent years, especially for elders who suffer from multiple diseases, such as hypertension, hyperlipidemia, cardiopathy, and cancers (Foucquier and Guedj, 2015; Li et al, 2015; MadaniTonekaboni et al, 2018)
Our NMDADNN method can be divided into four parts: 1) extracting drug features and computing similarity between drugs, 2) adjusting the drug similarity matrices by using random walk with restart algorithm and to compute positive pointwise mutual information for capturing the network structural information, 3) integrating the five drug similarity matrices with the multimodal deep autoencoder (MDA) method to obtain the unified embedding features for representing each drug, and 4) constructing the drug–drug pair{di, dj}features by concatenating the unified embedding features of drug diand drug dj and feeding them into the deep neural network (DNN) to predict the type of drug–drug interactions (DDIs) interaction
We first introduce six metrics and cross-validation test approaches to evaluate the performance of predictors and compare the performance of NMDADNN with other existing state-of-the-art DNN-based methods on the same data set, discussing the effect of Anatomical Therapeutic Chemical classification (ATC) feature, representation strategies, feature aggregate operators, and parameter setting
Summary
Combined drug therapies are becoming the prevalent approach for complex disease in recent years, especially for elders who suffer from multiple diseases, such as hypertension, hyperlipidemia, cardiopathy, and cancers (Foucquier and Guedj, 2015; Li et al, 2015; MadaniTonekaboni et al, 2018). Drug–drug interactions (DDIs) may cause adverse drug events (ADEs), reduce the efficacy, and so on. The synergistic reaction is the best result for combined drug therapies, meaning that the efficacy of drug A&B is bigger than the sum of drug A efficacy and drug B efficacy. The antagonistic reaction is the worst result for combined drug therapies, which results in reduced efficacy, and the efficacy of drug A&B is smaller than the sum of drug A efficacy and drug B efficacy. The type of no reaction is that the efficacy of drug A&B is equal to the sum of drug A efficacy and drug B efficacy; that is, there is no interaction between drugs A and B. Identifying the types of DDIs is very important in drug research, and it is helpful for safer and effective drug combined prescriptions, and may help in understanding the causes of side effects of existing drugs
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
