Data for: hERG Liability Classification Models Using Machine Learning Techniques

This file pertains 1) all SMILES(except the evaluation set-3 which contains the compound data from in-house proprietary projects) with respective pIC50 values that were used in training and evaluating the models 2) list of descriptors that were used to build models本文件包含如下两类内容：1）所有用于模型训练与评估的SMILES（Simplified Molecular Input Line Entry System，简化分子线性输入规范）及其对应的pIC50值，但评估集3除外；该评估集3包含来自内部专有项目的化合物数据。2）用于构建模型的分子描述符列表。