SCOPE-DTI
收藏资源简介:
SCOPE-DTI是一个大规模、平衡的半归纳人类药物-靶点相互作用数据集,由13个公共仓库构建而成,相比常见基准数据集,数据量扩大了100倍。该数据集整合了蛋白质和化合物的三维结构表示、图神经网络和双线性注意机制,有效捕捉跨领域交互模式,显著优于现有最先进方法。此外,SCOPE-DTI还提供了一个用户友好的界面和数据库,加速药物发现研究。
SCOPE-DTI is a large-scale, balanced semi-inductive human drug-target interaction (DTI) dataset constructed from 13 public repositories, with a scale 100 times larger than that of common benchmark DTI datasets. This dataset integrates three-dimensional structural representations of proteins and compounds, graph neural networks, and bilinear attention mechanisms, effectively capturing cross-domain interaction patterns and significantly outperforming existing state-of-the-art methods. Additionally, SCOPE-DTI provides a user-friendly interface and database to accelerate drug discovery research.

- 1SCOPE-DTI: Semi-Inductive Dataset Construction and Framework Optimization for Practical Usability Enhancement in Deep Learning-Based Drug Target Interaction Prediction香港中文大学(深圳)医学院 · 2025年



