SCOPE-DTI: Semi-Inductive Dataset Construction and Framework Optimization for Practical Usability Enhancement in Deep Learning-Based Drug Target Interaction Prediction

2025-03-12

AI Summary

To address the dual bottlenecks of insufficient real-world data diversity and excessive model complexity in drug–target interaction (DTI) prediction, this work proposes a semi-inductive modeling paradigm. First, we construct the first large-scale, class-balanced human DTI dataset incorporating 3D structural information—exceeding the Human benchmark in size by two orders of magnitude. Second, we design an end-to-end deep learning framework that jointly encodes 3D molecular and protein structures, integrates graph neural networks, and employs a bilinear attention mechanism to enable multi-source data fusion and semi-inductive inference. Our method achieves state-of-the-art performance across multiple evaluation metrics. We publicly release both the curated database and an interactive analytical platform. Furthermore, we experimentally validate the predicted anti-cancer target of ginsenoside Rh1, demonstrating practical utility in AI-driven drug discovery.

Technology Category

Humans and AI: Other Foundations of Human Computation & AIMachine Learning: Deep Neural Architectures and Foundation ModelsKnowledge Representation and Reasoning: Computational Complexity of Reasoning

Application Category

Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactionsGraph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphsEconomics, Online Markets and Human Computation: Data quality aspects of human-annotated datasets

Abstract

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. Constructed from 13 public repositories, the SCOPE dataset 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, significantly outperforming state-of-the-art methods across various DTI prediction tasks. Additionally, SCOPE-DTI provides a user-friendly interface and database. We further validate its effectiveness by experimentally identifying anticancer targets of Ginsenoside Rh1. By offering comprehensive data, advanced modeling, and accessible tools, SCOPE-DTI accelerates drug discovery research.

Problem

• Enhances real-world applicability of DTI prediction methods
• Addresses limited data diversity and modeling complexity
• Provides a user-friendly interface and database for drug discovery

Innovation

• Large-scale semi-inductive human DTI dataset
• 3D protein and compound representations integration
• Graph neural networks with bilinear attention