Abstract:Accurate identification of potential drug–target interactions (DTIs) is a crucial task in drug development and repositioning. Despite the remarkable progress achieved in recent years, improving the performance of DTI prediction still presents significant challenges. In this study, we propose a novel end-to-end deep learning model called AMMVF-DTI (attention mechanism and multi-view fusion), which leverages a multi-head self-attention mechanism to explore varying degrees of interaction between drugs and target proteins. More importantly, AMMVF-DTI extracts interactive features between drugs and proteins from both node-level and graph-level embeddings, enabling a more effective modeling of DTIs. This advantage is generally lacking in existing DTI prediction models. Consequently, when compared to many of the start-of-the-art methods, AMMVF-DTI demonstrated excellent performance on the human, C. elegans, and DrugBank baseline datasets, which can be attributed to its ability to incorporate interactive information and mine features from both local and global structures. The results from additional ablation experiments also confirmed the importance of each module in our AMMVF-DTI model. Finally, a case study is presented utilizing our model for COVID-19-related DTI prediction. We believe the AMMVF-DTI model can not only achieve reasonable accuracy in DTI prediction, but also provide insights into the understanding of potential interactions between drugs and targets.

What problem does this paper attempt to address?

The problem this paper attempts to address is the accurate identification of potential drug-target interactions (DTIs), which is a critical task in the process of drug development and repositioning. Despite significant progress in DTI prediction in recent years, improving DTI prediction performance still faces major challenges. Specifically, existing DTI prediction models often lack the ability to extract interaction information from both local and global structures, which limits their prediction accuracy. To address these issues, the authors propose a novel end-to-end deep learning model—AMMVF-DTI (Attention Mechanism and Multi-View Fusion-based model). This model utilizes a multi-head self-attention mechanism to explore different degrees of interaction between drugs and target proteins and extracts interaction features between drugs and proteins from both node-level and graph-level embeddings, thereby modeling DTIs more effectively. These advantages enable AMMVF-DTI to perform excellently on multiple benchmark datasets, particularly outperforming many existing state-of-the-art methods on human, C. elegans, and DrugBank datasets. Additionally, the importance of each module of the model was validated through ablation experiments, and a case study demonstrated the application effectiveness of AMMVF-DTI in COVID-19 related DTI prediction. In summary, this model not only improves the accuracy of DTI prediction but also provides valuable insights into the potential interactions between drugs and targets.

A Novel Model Predicting Drug–Target Interactions Based on Attention Mechanism and Multi-View Fusion