Prediction Of Drug-drug Interactions Research Articles

A physiologically-based pharmacokinetic/pharmacodynamic modeling approach for drug-drug-gene interaction evaluation of S-warfarin with fluconazole.

Warfarin is a widely used anticoagulant, and its S-enantiomer has higher potency compared to the R-enantiomer. S-warfarin is mainly metabolized by cytochrome P450 (CYP) 2C9, and its pharmacological target is vitamin K epoxide reductase complex subunit 1 (VKORC1). Both CYP2C9 and VKORC1 have genetic polymorphisms, leading to large variations in the pharmacokinetics (PKs) and pharmacodynamics (PDs) of warfarin in the population. This makes dosage management of warfarin difficult, especially in the case of drug-drug interactions (DDIs). This study provides a whole-body physiologically-based pharmacokinetic/PD (PBPK/PD) model of S-warfarin for predicting the effects of drug-drug-gene interactions on S-warfarin PKs and PDs. The PBPK/PD model of S-warfarin was developed in PK-Sim and MoBi. Drug-dependent parameters were obtained from the literature or optimized. Of the 34 S-warfarin plasma concentration-time profiles used, 96% predicted plasma concentrations within twofold range compared to observed data. For S-warfarin plasma concentration-time profiles with CYP2C9 genotype, 364 of 386 predicted plasma concentration values (~94%) fell within the twofold of the observed values. This model was tested in DDI predictions with fluconazole as CYP2C9 perpetrators, with all predicted DDI area under the plasma concentration-time curve to the last measurable timepoint (AUClast) ratio within twofold of the observed values. The anticoagulant effect of S-warfarin was described using an indirect response model, with all predicted international normalized ratio (INR) within twofold of the observed values. This model also incorporates a dose-adjustment method that can be used for dose adjustment and predict INR when warfarin is used in combination with CYP2C9 perpetrators.

Physiologically based pharmacokinetic modeling of imatinib and N-desmethyl imatinib for drug-drug interaction predictions.

The first-generation tyrosine kinase inhibitor imatinib has revolutionized the development of targeted cancer therapy and remains among the frontline treatments, for example, against chronic myeloid leukemia. As a substrate of cytochrome P450 (CYP) 2C8, CYP3A4, and various transporters, imatinib is highly susceptible to drug-drug interactions (DDIs) when co-administered with corresponding perpetrator drugs. Additionally, imatinib and its main metabolite N-desmethyl imatinib (NDMI) act as inhibitors of CYP2C8, CYP2D6, and CYP3A4 affecting their own metabolism as well as the exposure of co-medications. This work presents the development of a parent-metabolite whole-body physiologically based pharmacokinetic (PBPK) model for imatinib and NDMI used for the investigation and prediction of different DDI scenarios centered around imatinib as both a victim and perpetrator drug. Model development was performed in PK-Sim® using a total of 60 plasma concentration-time profiles of imatinib and NDMI in healthy subjects and cancer patients. Metabolism of both compounds was integrated via CYP2C8 and CYP3A4, with imatinib additionally transported via P-glycoprotein. The subsequently developed DDI network demonstrated good predictive performance. DDIs involving imatinib and NDMI were simulated with perpetrator drugs rifampicin, ketoconazole, and gemfibrozil as well as victim drugs simvastatin and metoprolol. Overall, 12/12 predicted DDI area under the curve determined between first and last plasma concentration measurements (AUClast) ratios and 12/12 predicted DDI maximum plasma concentration (Cmax) ratios were within twofold of the respective observed ratios. Potential applications of the final model include model-informed drug development or the support of model-informed precision dosing.

Quantitative prediction of CYP3A induction-mediated drug-drug interactions in clinical practice

There have been no reports on the quantitative prediction of CYP3A induction-mediated decreases in AUC and Cmax for drug candidates identified as a “victims” of CYP3A induction. Our previous study separately evaluated the fold-induction of hepatic and intestinal CYP3A by known inducers using clinical induction data and revealed that we were able to quantitatively predict the AUC ratio (AUCR) of a few CYP3A substrates in the presence and absence of CYP3A inducers. In the present study, we investigate the predictability of AUCR and also Cmax ratio (CmaxR) in additional 54 clinical studies. The fraction metabolized by CYP3A (fm), the intestinal bioavailability (Fg), and the hepatic intrinsic clearance (CLint) of substrates were determined by the in vitro experiments as well as clinical data used for calculating AUCR and CmaxR. The result showed that 65–69% and 65–67% of predictions were within 2-fold of observed AUCR and CmaxR, respectively. A simulation using multiple parameter combinations suggested that the variability of fm and Fg within a certain range might have a minimal impact on the calculation output. These findings suggest that clinical AUCR and CmaxR of CYP3A substrates can be quantitatively predicted from the preclinical stage.

Attention-Based Learning for Predicting Drug-Drug Interactions in Knowledge Graph Embedding Based on Multisource Fusion Information

Drug combinations can reduce drug resistance and side effects and enable the improvement of disease treatment efficacy. Therefore, how to effectively identify drug-drug interactions (DDIs) is a challenging problem. Currently, there exist several approaches that leverage advanced representation learning and graph-based techniques for DDIs prediction. While these methods have demonstrated promising results, a limited number of approaches effectively utilize the potential of knowledge graphs (KGs), which provide information on drug attributes and multirelation among entities. In this work, we introduce a novel attention-based KGs representation learning framework. To encode drug SMILES sequence, a pretrained model is used, while molecular structure information is mapped as the initialization of nodes within the KG using a message-passing neural network. Additionally, the knowledge-aware graph attention network is employed to capture the drug and its topological neighbor representation in the KG representation module. To prevent the oversmoothing problem, the residual layer is used in the DDI prediction module. Comprehensive experiments on several datasets have demonstrated that the proposed method outperforms the state-of-the-art algorithms on the DDI prediction task across a range of evaluation metrics. It achieves an accuracy of 0.924 and an AUC of 0.9705 on the KEGG dataset and attains an ACC of 0.9777 and an AUC of 0.9959 on the OGB-biokg dataset. These experimental findings affirm that our approach is a dependable model for predicting the association of drugs.

AutoDDI: Drug-Drug Interaction Prediction With Automated Graph Neural Network.

Drug-drug interaction (DDI) has attracted widespread attention because when incompatible drugs are taken together, DDI will lead to adverse effects on the body, such as drug poisoning or reduced drug efficacy. The adverse effects of DDI are closely determined by the molecular structures of the drugs involved. To represent drug data effectively, researchers usually treat the molecular structure of drugs as a molecule graph. Then, previous studies can use the handcrafted graph neural network (GNN) model to learn the molecular graph representations of drugs for DDI prediction. However, in the field of bioinformatics, manually designing GNN architectures for specific molecular structure datasets is time-consuming and depends on expert experience. To address this problem, we propose an automatic drug-drug interaction prediction method named AutoDDI that can efficiently and automatically design the GNN architecture for drug-drug interaction prediction without manual intervention. To this end, we first design an effective search space for drug-drug interaction prediction by revisiting various handcrafted GNN architectures. Then, to efficiently and automatically design the optimal GNN architecture for each drug dataset from the search space, a reinforcement learning search algorithm is adopted. The experiment results show that AutoDDI can achieve the best performance on two real-world datasets. Moreover, the visual interpretation results of the case study show that AutoDDI can effectively capture drug substructure for drug-drug interaction prediction.

Avapritinib Carries the Risk of Drug Interaction via Inhibition of UDP-Glucuronosyltransferase (UGT) 1A1.

Avapritinib is the only drug for adult patients with PDGFRA exon 18 mutated unresectable or metastatic gastrointestinal stromal tumor (GIST). Although avapritinib has been approved by the FDA for four years, little is known about the risk of drug-drug interactions (DDIs) via UDP-glucuronyltransferases (UGTs) inhibition. The aim of the present study was to systematically evaluate the inhibitory effects of avapritinib against UGTs and to quantitatively estimate its potential DDIs risk in vivo. Recombinant human UGTs were employed to catalyze the glucuronidation of substrates in a range of concentrations of avapritinib. The kinetics analysis was performed to evaluate the inhibition types of avapritinib against UGTs. The quantitative prediction of DDIs was done using in vitro-in vivo extrapolation (IVIVE). Avapritinib had a potent competitive inhibitory effect on UGT1A1. Quantitative prediction results showed that avapritinib administered at clinical doses might result in a 14.85% in-crease in area under the curve (AUC) of drugs primarily cleared by UGT1A1. Moreover, the Rgut value was calculated to be 18.44. Avapritinib has the potential to cause intestinal DDIs via the inhibition of UGT1A1. Additional attention should be paid when avapritinib is coadministered with UGT1A1 substrates.

DrugSimNet: Enhancing Drug Representation Learning through Drug Similarity for Accurate Drug-Drug Interaction Prediction

The field of drug discovery has garnered significant attention, particularly in light of advancements brought about by the "Homo Hundred" generation. Among the critical processes in drug discovery is drug screening, which is an important process in identifying and eliminating candidates that may pose potential side effects in the human body before in vivo experiments. The drug screening is often conducted via prediction of drug-drug Interaction, which utilizes algorithms to assess the likelihood and potential consequences of interactions between different drugs. This approach is a multifaceted process that involves the identification of potential new therapeutic entities by employing a combination of computational, experimental, translational, and clinical models. Traditional approaches to predicting drug-drug interactions in the context of drug discovery and polypharmacy heavily rely on empirical knowledge, in vitro assays, and animal experiments. However, these methods suffer from drawbacks such as being time-consuming, resource-intensive, and limited in their ability to capture the complete complexity of drug interactions. Therefore, there is a pressing need to develop automated and efficient methods that can accurately predict drug-drug interactions. To address the aforementioned problem, we proposed a novel representation learning based framework for prediction of drug-drug interaction. The proposed framework consists of two stages: representation learning, which focuses on extracting meaningful features from drugs, and transfer learning, utilized to train the drug-drug interaction prediction network. Through extensive experimentation, we have shown that the proposed drug-drug interaction prediction framework surpasses existing methods in terms of performance.

Enhancing Drug-Drug Interaction Prediction with Auxiliary Drug Similarity Estimation using Convolutional Neural Networks

In the last few years, a rapid development of artificial intelligence has revealed its possible applications in many tasks. One of those areas where artificial intelligence showed great promise was predicting drug-drug interactions, which had an enormous amount of data to be interpreted through. Drug-drug interactions are caused by interactions between drugs; it can decrease the effectiveness, lead to serious emergencies of the patient, or have synergies that increase the effectiveness of both. The previous knowledge-based drug-drug interaction prediction methods were costly and time consuming. Thus, a model that can predict these interactions is highly demanded for patients’ safety, furthermore, providing insights to scientists in medical fields. In this research, I propose a similarity aware drug-drug interaction predicting model. The proposed method consists of three modules: a drug feature extractor, a drug similarity estimator, and a drug-drug interaction predictor. The trained network is designed to be drug similarity-aware, which leads to significantly improved accuracy compared to the naive models. By incorporating drug similarity information, the model can provide more precise and reliable predictions for drug-drug interactions. The experimental result of the proposed deep-learning model achieved state-of-the-art performance compared to other previous methods.

Synergizing Convolutional Neural Networks and Drug Similarity Estimation for Improved Drug-Drug Interaction Prediction

Drug-drug interactions can cause adverse effects and impact patient safety. Traditionally, the prediction of DDIs has been labor intensive. However, these approaches have certain limitations in terms of scalability, convergence of different drug combinations, and capturing complex interactions. The potential combinations increase as the number of approved drugs increases; this results in the need for efficient methods. This research aims to utilize artificial intelligence and machine learning techniques to develop a method that accurately detects potential DDIs. The proposed idea allows for greater accuracy and automation in predicting DDI, reducing the time and cost required. The proposed method takes two drug formulas as input and employs a Drug Feature Extractor to extract their features. These features are then represented in feature maps, which are used when calculating a similarity score between input drugs and other drugs in the data set, and to predict potential DDIs. The model combines a drug similarity calculator and DDI predictor, enabling the system to process data in a “human-like” method; aiding in predicting interactions for newly developed drugs. The proposed model achieved state-of-art performance with an accuracy of 88.9%. The results demonstrate the efficacy of the proposed method in predicting potential drug interactions.

Artificial intelligence-driven drug interaction prediction

Artificial intelligence (AI) is developing at a rapid pace and this has led to revolutionary changes in many fields, including healthcare. Drug interaction prediction, which evaluates possible interactions between various medications to guarantee patient safety and maximize therapeutic outcomes is a crucial component of healthcare. This work investigates the use of artificial intelligence (AI) methods for predicting drug interactions, with a particular emphasis on the combination of natural language processing, knowledge graphs, and machine learning algorithms. The manual curation and experimental research that are frequently used in traditional drug interaction prediction methods limit their scalability and real-time applicability. On the other hand, artificial intelligence (AI) methods use molecular data, electronic health records, and large-scale healthcare data to improve the precision and effectiveness of drug interaction prediction. Deep neural networks and ensemble approaches are two examples of machine learning models that are essential for evaluating various datasets and spotting complex patterns related to drug interactions.

Prediction of multiple types of drug interactions based on multi-scale fusion and dual-view fusion.

Potential drug-drug interactions (DDI) can lead to adverse drug reactions (ADR), and DDI prediction can help pharmacy researchers detect harmful DDI early. However, existing DDI prediction methods fall short in fully capturing drug information. They typically employ a single-view input, focusing solely on drug features or drug networks. Moreover, they rely exclusively on the final model layer for predictions, overlooking the nuanced information present across various network layers. To address these limitations, we propose a multi-scale dual-view fusion (MSDF) method for DDI prediction. More specifically, MSDF first constructs two views, topological and feature views of drugs, as model inputs. Then a graph convolutional neural network is used to extract the feature representations from each view. On top of that, a multi-scale fusion module integrates information across different graph convolutional layers to create comprehensive drug embeddings. The embeddings from the two views are summed as the final representation for classification. Experiments on two real-world datasets demonstrate that MSDF achieves higher accuracy than state-of-the-art methods, as the dual-view, multi-scale approach better captures drug characteristics.

Critical role of growth medium for detecting drug interactions in Gram-negative bacteria that model in vivo responses.

Drug-resistant bacterial infections are a growing concern and have only continued to increase during the SARS-CoV-2 pandemic. Though not routinely used for Gram-negative bacteria, drug combinations are sometimes used for serious infections and may become more widely used as the prevalence of extremely drug-resistant organisms increases. To date, reliable methods are not available for identifying beneficial drug combinations for a particular infection. Our study shows variability across strains in how drug interactions are impacted by growth conditions. It also demonstrates that testing drug combinations in tissue-relevant growth conditions for some strains better models what happens during infection and may better inform combination therapy selection.

Physiologically Based Pharmacokinetic (PBPK) Modeling to Predict CYP3A-Mediated Drug Interaction between Saxagliptin and Nicardipine: Bridging Rat-to-Human Extrapolation.

The aim of this study was to predict the cytochrome P450 3A (CYP3A)-mediated drug-drug interactions (DDIs) between saxagliptin and nicardipine using a physiologically based pharmacokinetic (PBPK) model. Initially, in silico and in vitro parameters were gathered from experiments or the literature to construct PBPK models for each drug in rats. These models were integrated to predict the DDIs between saxagliptin, metabolized via CYP3A2, and nicardipine, exhibiting CYP3A inhibitory activity. The rat DDI PBPK model was completed by optimizing parameters using experimental rat plasma concentrations after co-administration of both drugs. Following co-administration in Sprague-Dawley rats, saxagliptin plasma concentration significantly increased, resulting in a 2.60-fold rise in AUC, accurately predicted by the rat PBPK model. Subsequently, the workflow of the rat PBPK model was applied to humans, creating a model capable of predicting DDIs between the two drugs in humans. Simulation from the human PBPK model indicated that nicardipine co-administration in humans resulted in a nearly unchanged AUC of saxagliptin, with an approximate 1.05-fold change, indicating no clinically significant changes and revealing a lack of direct translation of animal interaction results to humans. The animal-to-human PBPK model extrapolation used in this study could enhance the reliability of predicting drug interactions in clinical settings where DDI studies are challenging.

Deep learning for drug‐drug interaction prediction: A comprehensive review

AbstractThe prediction of drug‐drug interactions (DDIs) is a crucial task for drug safety research, and identifying potential DDIs helps us to explore the mechanism behind combinatorial therapy. Traditional wet chemical experiments for DDI are cumbersome and time‐consuming, and are too small in scale, limiting the efficiency of DDI predictions. Therefore, it is particularly crucial to develop improved computational methods for detecting drug interactions. With the development of deep learning, several computational models based on deep learning have been proposed for DDI prediction. In this review, we summarized the high‐quality DDI prediction methods based on deep learning in recent years, and divided them into four categories: neural network‐based methods, graph neural network‐based methods, knowledge graph‐based methods, and multimodal‐based methods. Furthermore, we discuss the challenges of existing methods and future potential perspectives. This review reveals that deep learning can significantly improve DDI prediction performance compared to traditional machine learning. Deep learning models can scale to large‐scale datasets and accept multiple data types as input, thus making DDI predictions more efficient and accurate.

Physiologically Based Pharmacokinetic (PBPK) Modeling of small molecules: How Much Progress Have We Made?

Physiologically based pharmacokinetic (PBPK) models of small molecules have become mainstream in drug development and academic research. The use of PBPK models is continuously expanding with the majority of work now focusing on predictions of drug-drug interactions, drug-disease interactions, and changes in drug disposition across lifespan. Recently, publications that use PBPK modeling to predict drug disposition during pregnancy and in organ impairment have increased reflecting the advances in incorporating diverse physiological changes into the models. Due to the expanding computational power and diversity of modeling platforms available, the complexity of PBPK models has also increased. Academic efforts have provided clear advances in better capturing human physiology in PBPK models and incorporating more complex mathematical concepts into PBPK models. Examples of such advances include the segregated gut model with a series gut compartments allowing modeling of physiological blood flow distribution within an organ and zonation of metabolic enzymes, and series compartment liver models allowing simulations of hepatic clearance for high extraction drugs. Despite these advances in academic research, the progress in assessing model quality and defining model acceptance criteria based on the intended use of the models has not kept pace. This review suggests that awareness of the need for predefined criteria for model acceptance has increased but many manuscripts still lack description of scientific justification and/or rationale for chosen acceptance criteria. As artificial intelligence and machine learning approaches become more broadly accepted, these tools offer promise for development of comprehensive assessment for existing observed data and analysis of model performance. Significance Statement PBPK modeling has become a mainstream application in academic literature and is broadly used for predictions, analysis and evaluation of pharmacokinetic data. Many significant advances have been made in developing advanced PBPK models that better capture human physiology but oftentimes sufficient justification for the chosen model acceptance criterion and model structure is still missing. This review provides a summary of the current landscape of PBPK applications used and highlights the needs for advancing PBPK modeling science and training in academia.

Use of a Double-Transfected System to Predict hOCT2/hMATE1-Mediated Renal Drug-Drug Interactions.

Accurate predictions of renal drug-drug interactions (DDIs) mediated by the human organic cation transporter 2 (hOCT2) and multidrug and toxin extrusion proteins (hMATEs) remain challenging. Current DDI evaluation using plasma maximal unbound inhibitor concentrations (Imax,u) and IC50 values determined in single transporter-transfected cells frequently leads to false or overprediction especially for hMATE1. Emerging evidence suggests intracellular unbound inhibitor concentration may be more relevant for hMATE1 inhibition in vivo. However, determination of intrarenal inhibitor concentrations is impractical. Here, we explored the use of hOCT2/hMATE1 double-transfected Madin-Darby canine kidney (MDCK) cells as a new in vitro tool for DDI risk assessment. Our results showed that potent in vitro hMATE1 inhibitors (hydroxychloroquine, brigatinib, and famotidine) failed to inhibit metformin B-to-A flux in the double-transfected system. On the other side, the classic hOCT2/hMATE1 inhibitors, pyrimethamine and cimetidine, dose-dependently inhibited metformin apparent B-to-A permeability (Papp). The different behaviors of these hMATE1 inhibitors in the double-transfected system can be explained by their different ability to gain intracellular access either via passive diffusion or transporter-mediated uptake. A new parameter (IC50,flux) was proposed reflecting the inhibitor's potency on overall hOCT2/hMATE1-mediated tubular secretion. The IC50,flux values significantly differ from the IC50 values determined in single transporter-transfected cells. Importantly, the IC50,flux accurately predicted in vivo DDIs (within 2-fold) when used in a static model. Our data demonstrated that the IC50,flux approach circumvents the need to measure intracellular inhibitor concentrations and more accurately predicted hOCT2/hMATE1-mediated renal DDIs. This system represents a new approach that could be used for improved DDI assessment during drug development. SIGNIFICANCE STATEMENT: This study demonstrated that flux studies in double-transfected MDCK cells and the IC50,flux represents a better approach to assess in vivo DDI potential for the renal organic cation secretion system. This study highlights the importance of inhibitor intracellular accessibility for accurate prediction of hMATE1-mediated renal DDIs. This approach has the potential to identify in vitro hMATE1 inhibitors that are unlikely to result in in vivo DDIs, thus reducing the burden of unnecessary and costly clinical DDI investigations.

Retrospective Analysis of Potential Adverse Drug Interactions in the Drugs Prescribed to the Elderly at a Tertiary Health Care Center in Raipur, Chhattisgarh, Central India.

Background The elderly population differs from adults in having various physiological changes and multiple diseases, which demand the use of multiple medications. The practice of polypharmacy in the elderly leads to numerous harmful effects like adverse drug reactions, adverse drug-drug interactions (DDIs), poor compliance, etc. Methodology This study collected 295 case files of elderly patients retrospectively in the Departments of General Medicine, Cardiology and Nephrology after obtaining Institute Ethics Committee approval to look for the potential adverse DDIs with their severity according to the clinical significance. Results and interpretation The total numberof adverse DDIs identified was 156, the maximum in Category 'C.' Salbutamol plus carvedilol/propranolol, ramipril plus telmisartan and ivabradine plus diltiazem were the adverse DDIs categorized under severity 'X'. The identified DDIs were categorised according to the mechanism such as increased bleeding risk, hypokalemia, hyperkalaemia, reduced effect of drugs and increased effect of drugs. Conclusion Polypharmacy can lead to several adverse consequences in the elderly, of whichadverse DDIs play a crucial role in harmful health outcomes. This study brings out the significance of predicting drug interactions beforehand which can reduce the risk of bleeding and other risks of hyper/hypokalaemia, hyponatremia and hypoglycaemia.

Learning self-supervised molecular representations for drug–drug interaction prediction

Drug–drug interactions (DDI) are a critical concern in healthcare due to their potential to cause adverse effects and compromise patient safety. Supervised machine learning models for DDI prediction need to be optimized to learn abstract, transferable features, and generalize to larger chemical spaces, primarily due to the scarcity of high-quality labeled DDI data. Inspired by recent advances in computer vision, we present SMR–DDI, a self-supervised framework that leverages contrastive learning to embed drugs into a scaffold-based feature space. Molecular scaffolds represent the core structural motifs that drive pharmacological activities, making them valuable for learning informative representations. Specifically, we pre-trained SMR–DDI on a large-scale unlabeled molecular dataset. We generated augmented views for each molecule via SMILES enumeration and optimized the embedding process through contrastive loss minimization between views. This enables the model to capture relevant and robust molecular features while reducing noise. We then transfer the learned representations for the downstream prediction of DDI. Experiments show that the new feature space has comparable expressivity to state-of-the-art molecular representations and achieved competitive DDI prediction results while training on less data. Additional investigations also revealed that pre-training on more extensive and diverse unlabeled molecular datasets improved the model’s capability to embed molecules more effectively. Our results highlight contrastive learning as a promising approach for DDI prediction that can identify potentially hazardous drug combinations using only structural information.

MFD–GDrug: multimodal feature fusion-based deep learning for GPCR–drug interaction prediction

The accurate identification of drug–protein interactions (DPIs) is crucial in drug development, especially concerning G protein-coupled receptors (GPCRs), which are vital targets in drug discovery. However, experimental validation of GPCR–drug pairings is costly, prompting the need for accurate predictive methods. To address this, we propose MFD–GDrug, a multimodal deep learning model. Leveraging the ESM pretrained model, we extract protein features and employ a CNN for protein feature representation. For drugs, we integrated multimodal features of drug molecular structures, including three-dimensional features derived from Mol2vec and the topological information of drug graph structures extracted through Graph Convolutional Neural Networks (GCN). By combining structural characterizations and pretrained embeddings, our model effectively captures GPCR–drug interactions. Our tests on leading GPCR–drug interaction datasets show that MFD–GDrug outperforms other methods, demonstrating superior predictive accuracy.

MAVGAE: a multimodal framework for predicting asymmetric drug–drug interactions based on variational graph autoencoder

Drug-drug interactions refer to the phenomena wherein the potency, duration, or effectiveness of one or multiple drugs undergo alterations of varying degrees as a result of their concurrent or sequential usage. The accurate identification of potential drug interactions plays a pivotal role in mitigating the risks associated with drug administration in patients, it also helps in minimizing the likelihood of hazardous situations arising during a patient’s course of treatment. However, researchers have found that there is a problem of asymmetric drug interactions, where one drug may affect another but not vice versa. This adds to the difficulty of prediction, so in polypharmacy, the order of drug administration is critical to efficacy and safety, and few current studies predict asymmetric DDIs. Aiming at the above problems, we propose a framework based on multimodal data and a variational graph autoencoder named MAVGAE for predicting asymmetric drug interactions. The framework initially encodes multimodal data into low-dimensional representations and then utilizes a variational graph autoencoder for encoding and decoding. During the model training process, supervised learning is employed for the classification task with the incorporation of heterogeneity information, ensuring accurate prediction of drug interactions. Experimental validation on a large-scale drug dataset demonstrates the framework’s high accuracy and reliability in predicting non-symmetrical drug interactions, offering effective support and guidance for drug research.