[Hilal Tayara] Developing DDINet For Accurate and Scalable Drug-Drug Interaction Prediction
The proposed deep learning model can accurately predict drug-drug interactions for even new, unseen drugs, representing a new paradigm Drug-drug interactions (DDI) can cause adverse drug reactions during the co-administration of multiple drugs, necessitating accurate and scalable prediction tools. While deep learning models have shown promise recently, most models show poor performance against drugs not encountered during training. Now, researchers have developed a lightweight and scalable model, called DDINet, designed specifically to predict unseen drug interactions. This innovative model achieves superior accuracy in predicting interactions for unseen drugs, with potential for practical deployment. Title: DDINet Architecture Caption: DDINet utilizes a streamlined deep learning architecture, while being lightweight and scalable. It demonstrates excellent performance in predicting interaction of new, unseen drugs. Credit: Associate Professor Hilal Tayara from Jeonbuk National University License type: Original Content Usage restrictions: Cannot be reused without permission Managing complex medical conditions often requires the simultaneous use of multiple different drugs, referred to as polypharmacy. While necessary, this significantly increases the risk of drug–drug interactions (DDIs), which can either enhance or decrease therapeutic effects or trigger adverse drug reactions (ADRs), potentially leading to longer hospital stays or even life-threatening outcomes. In recent years, researchers have increasingly turned to deep learning models to predict DDIs. Although these models often outperform traditional methods, they are usually tested under idealized conditions, in which training and test data are randomly split, failing to reflect real-world clinical settings. As a result, many existing models suffer sharp drops in performance when evaluated on truly unseen drugs. Some also require substantial computational resources, limiting real-world usability. To overcome these limitations, a research team led by Associate Professor Hilal Tayara from the School of International Engineering and Science at Jeonbuk National University (JBNU), South Korea, has developed DDINet, a lightweight and scalable model, specifically designed to predict interactions for new, unseen drugs. “DDINet can simultaneously predict whether an interaction will occur and identify its biological effect, while needing significantly less computational power than complex graph-based models,” explains Dr. Tayara. Their study was made available online on November 29, 2025, and published in Volume 333 of Knowledge-Based Systems on January 30, 2026. DDINet utilizes a streamlined architecture with five fully connected layers and uses molecular fingerprints of drugs as input. This approach avoids overfitting to training data – a common reason why many models struggle to generalize to unseen drugs. Importantly, it is designed to handle binary classification tasks, which involve predicating the likelihood of whether a given drug pair will interact, and multi-classification tasks, where the goal is to predict the biological effect or mechanisms of a known DDI. The researchers trained and evaluated DDINet using a large-scale dataset constructed from DrugBank. They also tested five different molecular fingerprinting techniques. To achieve enhanced generalization, the researchers adopted a strict data-splitting protocol during evaluation. Specifically, they created three scenarios for model evaluation. In scenario one (S1), drug pairs were randomly split into training and test datasets. Further, they utilized a DDI-based splitting where 10% of all DDI pairs formed an independent test set, and the remaining were used for training. Scenario two included DDIs where one drug was known and another unseen, while scenario three comprised DDIs where both drugs were unseen, representing realistic clinical settings. To categorize drugs as unseen and seen, the team applied a strict drug-based splitting protocol based on DrugBank annotations. Morgan fingerprints were identified as the best performing and were used for the final implementation. Across all evaluation scenarios, DDINet performed as well as or better than existing models, particularly in the most difficult S3. It demonstrated stable performance across a range of metrics in both binary and multi-classification tasks. “DDINet’s compact and efficient architecture enables large-scale deployment in hospitals, drug discovery pipelines, and pharmacovigilance systems,” concludes Dr. Tayara. “Ultimately, this technology can help accelerate drug development, while improving the safety of patients who rely on multiple medications.” [Reference] Title of original paper: DDINet: A multi-task neural network for accurate drug-drug interaction prediction and effect analysis Journal: Knowledge-Based Systems DOI:10.1016/j.knosys.2025.114981 [About the authors] PURE Author Profile Dr. Hilal Tayara is an Associate Professor in the School of International Engineering and Science at Jeonbuk National University (JBNU), South Korea. He holds a Ph.D. in Electronics and Information Engineering, specializing in Artificial Intelligence applications in Bioinformatics and drug discovery. His research group focuses on developing deep learning architectures for genomic analysis and computational drug safety. Prof. Tayara has published over 100 academic papers and was featured in Stanford University’s World’s Top 2% Scientists List (2022–2025). He received the Excellent Research Award from JBNU in October 2021 and was named a JBNU Fellow in December 2025. Sabir Ali received the M.Sc. degree in Information Technology from Quaid-e-Azam University, Islamabad, Pakistan, in 2019. He is currently pursuing an integrated M.S.–Ph.D. degree in the Department of Electronics and Information Engineering at Jeonbuk National University, Jeonju, South Korea. His research interests include artificial intelligence, machine learning, transformer-based models, computational drug discovery, drug–drug interaction (DDI) prediction, adverse effect analysis, medical imaging, and biomedical data analysis. Dr. Waleed Alam received his Ph.D. in Electronics and Information Engineering from Jeonbuk National University, South Korea, in 2024. He is currently a Postdoctoral Fellow at the Chinese Institute of Brain Research, where his research focuses on artificial intelligence for brain imaging, as well as bioinformatics, computational biology, and computational drug discovery. He has published more than eight papers in drug discovery and bioinformatics and is currently working on brain image analysis. Prof. Kil To Chong is CEO of Juyoungbio LTD and Professor at the School of Electrical Engineering, Jeonbuk National University (since 1995), with leadership roles including Head of the Research Institute of Advanced Electronics and Information Technology and former Dean of the Global Frontier College.
KOR
