Predicting adverse drug reactions of combined medication from heterogeneous pharmacologic databases

Abstract

BackgroundEarly and accurate identification of potential adverse drug reactions (ADRs) for combined medication is vital for public health. Existing methods either rely on expensive wet-lab experiments or detecting existing associations from related records. Thus, they inevitably suffer under-reporting, delays in reporting, and inability to detect ADRs for new and rare drugs. The current application of machine learning methods is severely impeded by the lack of proper drug representation and credible negative samples. Therefore, a method to represent drugs properly and to select credible negative samples becomes vital in applying machine learning methods to this problem.ResultsIn this work, we propose a machine learning method to predict ADRs of combined medication from pharmacologic databases by building up highly-credible negative samples (HCNS-ADR). Specifically, we fuse heterogeneous information from different databases and represent each drug as a multi-dimensional vector according to its chemical substructures, target proteins, substituents, and related pathways first. Then, a drug-pair vector is obtained by appending the vector of one drug to the other. Next, we construct a drug-disease-gene network and devise a scoring method to measure the interaction probability of every drug pair via network analysis. Drug pairs with lower interaction probability are preferentially selected as negative samples. Following that, the validated positive samples and the selected credible negative samples are projected into a lower-dimensional space using the principal component analysis. Finally, a classifier is built for each ADR using its positive and negative samples with reduced dimensions. The performance of the proposed method is evaluated on simulative prediction for 1276 ADRs and 1048 drugs, comparing using four machine learning algorithms and with two baseline approaches. Extensive experiments show that the proposed way to represent drugs characterizes drugs accurately. With highly-credible negative samples selected by HCNS-ADR, the four machine learning algorithms achieve significant performance improvements. HCNS-ADR is also shown to be able to predict both known and novel drug-drug-ADR associations, outperforming two other baseline approaches significantly.ConclusionsThe results demonstrate that integration of different drug properties to represent drugs are valuable for ADR prediction of combined medication and the selection of highly-credible negative samples can significantly improve the prediction performance.Electronic supplementary materialThe online version of this article (10.1186/s12859-018-2520-8) contains supplementary material, which is available to authorized users.

Database of adverse events associated with drugs and drug combinations

“…Several methods have been published that aim to predict DDI induced adverse events 23,24 . Our method is different in spirit, as it seeks to extract (rather than predict with machine learning) data on adverse events.…”

Predicting adverse drug reactions of two‐drug combinations using structural and transcriptomic drug representations to train an artificial neural network

“…The differentiation is either in terms of a completely new type or a significant enhancement of an existing adverse event when compared to what is caused by a single drug. In effect, in this study and others (Liu et al, 2017;Zheng et al, 2018) the TWOSIDES dataset is used as a source of verified ADRs for combination of two drugs.…”

Trader as a new optimization algorithm predicts drug-target interactions efficiently

“…In a series of related works, the designed network is examined by various algorithms such as Random walk 16,17 and Random forest 18 . Unlike the first class of related works which depends on the negative dataset 19 , the second group only considers the existing information. As a result, the error of the second category is lower than the first one.…”