CandidateDrug4Cancer

CandidateDrug4Cancer是由平安健康科技北京有限公司创建的一个开放的分子图学习基准数据集，专注于癌症药物发现。该数据集包含54869个与癌症相关的药物分子，涵盖了从临床前、临床到FDA批准的不同阶段。数据集通过整合ChEMBL数据库中的信息创建，旨在通过图机器学习研究，促进可扩展、稳健和可重复的抗癌症药物发现。该数据集的应用领域主要集中在通过计算机辅助方法预测药物与目标的相互作用，以加速新药的开发。

CandidateDrug4Cancer is an open molecular graph learning benchmark dataset focused on cancer drug discovery, developed by Ping An Health Technology Beijing Co., Ltd. It contains 54,869 cancer-associated drug molecules, covering diverse stages ranging from preclinical, clinical trials to FDA-approved agents. Constructed by integrating data from the ChEMBL database, this dataset aims to facilitate scalable, robust and reproducible anti-cancer drug discovery via graph machine learning research. Its primary application scenarios focus on predicting drug-target interactions through computer-aided methods to accelerate new drug development.