Unraveling the origin of splice switching activity of hemoglobin beta-globin gene modulators via QSAR modeling

This supplementary data is comprised of 3 files as follows: 1. active.zip (CSV of the set of active compounds) 2. inactive.zip (CSV of the set of inactive compounds) 3. 50models.zip (CSV for 50 models derived from independent data splits) Note: Each CSV file are comprised of the SMILES, CID, Activity class label for 50models.zip as well as the selected features.

本补充数据集包含以下3个文件：1. active.zip（活性化合物集合对应的CSV文件）；2. inactive.zip（非活性化合物集合对应的CSV文件）；3. 50models.zip（基于独立数据拆分得到的50个模型对应的CSV文件）。注：50models.zip所包含的CSV文件中，包含简化分子线性输入规范（SMILES）、化合物标识符（CID）、活性类别标签以及选定的特征。