Semi-inductive dataset construction and framework optimization for practical drug target interaction prediction with ScopeDTI

Deep learning-based drug-target interaction (DTI) prediction methods have demonstrated strong performance; however, real-world applicability remains constrained by limited data diversity and modeling complexity. To address these challenges, we propose SCOPE-DTI, a unified framework combining a large-scale, balanced semi-inductive human DTI dataset with advanced deep learning modeling. SCOPE-DTI is constructed from 13 public repositories and expands data volume by up to 100-fold compared to common benchmarks such as the Human dataset. The SCOPE model integrates three-dimensional protein and compound representations, graph neural networks, and bilinear attention mechanisms to effectively capture cross domain interaction patterns and outperform state-of-the-art methods across various DTI prediction tasks. Additionally, SCOPE-DTI provides a user-friendly interface and database. We further demonstrate its effectiveness by experimentally identifying anticancer targets of two bioactive natural compounds. By offering comprehensive data, advanced modeling, and accessible tools, SCOPE-DTI accelerates drug discovery research. SCOPE-DTI introduces a large semi-inductive DTI dataset and optimized deep learning framework for drug-target interaction prediction. It integrates 3D structural modeling and bilinear attention, with experimental validation of drug targets and open access tools for broad research use.