Author & Research Contributor
Published in 2026 | VOLUME 03, JUNE ISSUE 06Graph Neural Networks (GNNs) are deep learning models that process graph-structured data, which consists of nodes and edges. They capture complex relationships and interaction, making them ideal for tasks like molecular drug discovery, Drug Target Interactions (DTI), and Drug Target (Binding) Affinity (DTA). GNNs revolutionize drug discovery by treating molecules as graphs, with atoms as nodes and chemical bonds as edges. This allows models to understand the complex three-dimensional structures and patterns relevant to biological activity. In poly-pharmacology, GNNs effectively model drug-drug interactions (DDIs) and predict multi-target activities. They improve predictive accuracy, lower development costs, and reduce late-stage failures. This review examines the application of GNNs in various phases of drug discovery, including lead discovery, optimization, synthetic route design, drug-target interaction prediction, and molecular property profiling, while addressing challenges in translational medicine. Computational drug-target affinity prediction has the potential to accelerate drug discovery. Currently, pre-training models have achieved significant success in various fields due to their ability to train the model using vast amounts of unlabelled data. However, given the scarcity of drug-target interaction data, pre-training models can only be trained separately on drug and target data, resulting in features that are insufficient for drug-target affinity prediction. To address this issue, a graph neural pre-training-based drug-target affinity prediction method (GNPDTA). This approach comprises three stages. In the first stage, two pre-training models are utilized to extract low-level features from drug atom graphs and target residue graphs, leveraging a large number of unlabelled training samples. In the second stage, two 2D convolutional neural networks are employed to combine the extracted drug atom features and target residue features into high-level representations of drugs and targets. Finally, in the third stage, a predictor is used to predict the drug-target affinity. This approach fully utilizes both unlabelled and labelled training samples, enhancing the effectiveness of pre-training models for drug-target affinity prediction. In our experiments, GNPDTA outperforms other deep learning methods, validating the efficacy of our approaches.
Graph neural networks; Lead discovery; Lead optimization; Synthetic route; Drug—target interaction; Property prediction; Virtual screening; De novo drug design, drug-target affinity, pre-training model, graph isomorphism network, deep neural network, feature extraction.