scbirlab/lyu-wang-balius-singh-2019-d4

该数据集来源于John J. Irwin, Bryan L. Roth, 和 Brian K. Shoichet的实验室，主要用于化学属性预测。数据集包含约1.15亿个化合物的SMILES字符串，并标注了ZINC IDs、重原子计数（HAC）和DOCKscore。此外，还使用RDKit计算了分子量、Crippen cLogP和拓扑表面积。数据集的创建是为了研究增加虚拟库中化合物的数量是否会增加在湿实验室中验证的新活性化学支架的数量。

This dataset originates from a massive docking campaign conducted by Lyu et al. (2019) to explore the potential of increasing the number of compounds in virtual libraries to discover new active chemical scaffolds. The dataset contains compounds represented as SMILES strings, annotated with ZINC IDs, heavy atom count (HAC), and DOCKscore. Additional molecular properties such as molecular weight, Crippen cLogP, and topological surface area were calculated using RDKit via the schemist tool. The dataset focuses on compounds docked to the D4 dopamine receptor, with compounds having anomalous DOCKscores removed. The dataset is licensed under CC-BY-4.0.