Predicting drug-target interaction using a novel graph neural network with 3D structure-embedded graph representation

Abstract

Accurate prediction of drug-target interaction (DTI) is essential for reliable in-silico drug design. Recently, deep learning techniques with exponentially growing bioactivity data are attracting considerable attention as a promising alternative for predicting DTI. Accordingly, we propose a novel deep learning approach for predicting drug-target interaction using a graph neural network. Graph neural networks show remarkable performance in predicting various molecular properties. Specifically, we introduce a distance-aware graph attention algorithm to differentiate various types of intermolecular interactions. Furthermore, we extract the graph feature of intermolecular interactions directly from the 3D structural information on the protein-ligand binding pose. Thus, the model can learn key features for accurate predictions of drug-target interaction rather than just memorize specific patterns of ligand molecules. As a result, our model shows better performance than docking and other deep learning methods for both virtual screening (AUROC of 0.968 for the DUD-E test set) and pose prediction (AUROC of 0.935 for the PDBbind test set). Also, it can reproduce the natural population distribution of active molecules and inactive molecules. Finally, we analyzed the common limitations of our model and other deep learning-based DTI models in generalization ability.