DynamicBind 蛋白质动态复合物结构预测数据集

该数据集是论文「DynamicBind: predicting ligand-specific protein-ligand complex structure with a deep equivariant generative model」的训练集和测试集，该论文是上海交通大学郑双佳课题组联合星药科技、中山大学药学院以及美国莱斯大学发布的，研究团队提出了为蛋白质「动态对接」设计的几何深度生成模型 DynamicBind ， 可以有效地将蛋白质构象从最初的 AlphaFold 预测调整到类似全息的状态，为后 AlphaFold 时代的药物研发提供了一种基于深度学习的、考虑蛋白动态变化的新研究范式。

This dataset comprises the training and test sets from the paper "DynamicBind: Predicting Ligand-Specific Protein-Ligand Complex Structure with a Deep Equivariant Generative Model", which was published by the research team led by Professor Zheng Shuangjia from Shanghai Jiao Tong University, in collaboration with Starpharma, the School of Pharmaceutical Sciences of Sun Yat-sen University, and Rice University of the United States. The research team proposed the geometric deep generative model DynamicBind designed for protein "dynamic docking", which can effectively adjust protein conformations from initially AlphaFold-predicted structures to a holographic-like state, providing a deep learning-based novel research paradigm that accounts for protein dynamic changes for drug discovery in the post-AlphaFold era.

搜集汇总

背景与挑战

背景概述

该数据集为论文《DynamicBind: predicting ligand-specific protein-ligand complex structure with a deep equivariant generative model》的训练和测试集，用于蛋白质动态复合物结构预测研究。数据基于PDbbind2020公共数据集处理而成，支持考虑蛋白质动态变化的深度生成模型应用。