Drug–drug interaction prediction: databases, web servers and computational models

Abstract

In clinical treatment, two or more drugs (i.e. drug combination) are simultaneously or successively used for therapy with the purpose of primarily enhancing the therapeutic efficacy or reducing drug side effects. However, inappropriate drug combination may not only fail to improve efficacy, but even lead to adverse reactions. Therefore, according to the basic principle of improving the efficacy and/or reducing adverse reactions, we should study drug–drug interactions (DDIs) comprehensively and thoroughly so as to reasonably use drug combination. In this review, we first introduced the basic conception and classification of DDIs. Further, some important publicly available databases and web servers about experimentally verified or predicted DDIs were briefly described. As an effective auxiliary tool, computational models for predicting DDIs can not only save the cost of biological experiments, but also provide relevant guidance for combination therapy to some extent. Therefore, we summarized three types of prediction models (including traditional machine learning-based models, deep learning-based models and score function-based models) proposed during recent years and discussed the advantages as well as limitations of them. Besides, we pointed out the problems that need to be solved in the future research of DDIs prediction and provided corresponding suggestions.

Detection of muscular system adverse reaction signals in sacubitril/valsartan treatment combined with statins

“…Investigations of DDI are limited to pre-market clinical trials of pharmacokinetics, and it is impossible to evaluate all possible interactions of a new drug comprehensively ( Zhao et al, 2023 ). Therefore, many DDI signals are not discovered until the drug is put on the market.…”

Predicting the physiological effects of multiple drugs using electronic health record

“…To optimize the performance of the XGBoost models, the Bayesian optimization approach was employed using Optuna 34 . Following hyperparameter were examined for the designated ranges in the Bayesian optimization: learning_rate, 10 -4 to 10 -1 , max_depth, 3 to 10, min_child_weight, 10 -3 to 10 2 ), subsample, 0.1 to 1, colsample_bytree, 0.1 to 1, reg_alpha, 10 -6 to 10 2 , reg_lambda, 10 -6 to 10 2 , and n_estimators, 50 to 300. During the Bayesian optimization process, an early stopping criterion was set to 20 rounds.…”

“…Consequently, understanding the effects of multiple drugs has become critical. In recent years, machine learning models have been developed to predict the effects of taking multiple drugs, often drug-drug interactions (DDIs), by processing various types of information 2,3,4 . DeepDDI 5 and DeepDDI2 6 use SMILES of two input drugs to predict their DDI effects by using deep learning.…”