PDCs数据集

PDCs数据集是由华南理工大学等机构构建的，用于探索肽-药物偶联物（PDCs）的结构-活性关系和活性预测的基准数据集。该数据集包含834个抗癌PDCs的结构和活性标签，涵盖了广泛的结构多样性和化学空间。数据集的构建过程严谨，每个条目包含三个关键组成部分：肽序列、连接剂SMILES和载荷SMILES。该数据集旨在解决PDCs设计的复杂性和缺乏系统性预测工具的问题，为开发新的PDCs药物提供了重要的数据基础。

The PDCs dataset is a benchmark dataset developed by institutions including South China University of Technology for exploring the structure-activity relationship (SAR) and activity prediction of peptide-drug conjugates (PDCs). It contains structural information and activity labels for 834 anticancer PDCs, spanning a broad spectrum of structural diversity and chemical space. The dataset was constructed rigorously, with each entry comprising three core components: peptide sequence, linker SMILES, and payload SMILES. This dataset aims to resolve the complexity of PDC design and the scarcity of systematic prediction tools, serving as a critical data foundation for the development of novel PDC-based therapeutics.