SidechainNet

SidechainNet是由匹兹堡大学开发的蛋白质序列和结构数据集，旨在解决蛋白质结构预测中侧链信息缺失的问题。该数据集扩展自ProteinNet，包含了描述蛋白质结构中所有重原子的角度和原子坐标信息。数据集的创建过程涉及从Protein Data Bank重新下载所有原子蛋白质结构，并整合了多种数据类型。SidechainNet的应用领域包括结构生物学、药物发现和分子动力学模拟，旨在提高蛋白质结构预测的准确性和表达能力。

SidechainNet is a protein sequence and structure dataset developed by the University of Pittsburgh, designed to address the problem of missing side chain information in protein structure prediction. This dataset is extended from ProteinNet, and contains angle and atomic coordinate information for all heavy atoms in protein structures. The dataset creation process involves re-downloading all atomic protein structures from the Protein Data Bank and integrating multiple data types. The application fields of SidechainNet include structural biology, drug discovery and molecular dynamics simulations, aiming to improve the accuracy and expressive capability of protein structure prediction.