QSAR-DATASET-FOR-DRUG-TARGET-CHEMBL3980

This dataset contains QSAR data (from ChEMBL version 17) showing activity values (unit is pseudo-pCI50) of several compounds on drug target ChEMBL_ID: CHEMBL3980 (TID: 30005), and it has 81 rows and 520 features (not including molecule IDs and class feature: molecule_id and pXC50). The features represent Molecular Descriptors which were generated from SMILES strings. Missing value imputation was applied to this dataset (By choosing the Median). Feature selection was also applied.

本数据集包含源自ChEMBL数据库17版的定量构效关系（Quantitative Structure-Activity Relationship, QSAR）数据，记录了多种化合物针对药物靶点ChEMBL_ID: CHEMBL3980（靶点ID（TID）: 30005）的活性值，活性值单位为伪pCI50。该数据集共包含81条样本与520个特征（不含分子ID与类别特征：molecule_id及pXC50）。所述特征为由简化分子线性输入规范（Simplified Molecular Input Line Entry System, SMILES）字符串生成的分子描述符。本数据集已完成缺失值处理，采用中位数法进行缺失值插补，同时亦开展了特征选择操作。