Explainable No-Code OECD-Compliant Machine Learning Models to Predict the Mutagenic Activity of Polycyclic Aromatic Hydrocarbons and Their Radical Cation Metabolites

Supplementary Material for "Explainable No-Code OECD-Compliant Machine Learning Models to Predict the Mutagenic Activity of Polycyclic Aromatic Hydrocarbons and Their Radical Cation Metabolites" (Submitted) : All supporting files necessary for this research are combined in a single .ZIP archive. The archive includes two primary folders: 'Outputs' and 'WEKA File', holding .xyz files for the molecular structures of the activated metabolites and procarcinogens, respectively. The 'WEKA File’ folder also provides the .arff files used for training and testing, organized by each specific data split. The ‘PAHs-GFN2-Data.xlsx' file includes the primary datasets used in this study, including all extracted molecular details, SMILES notations, numerical IDs, CDFT descriptors, and biological responses, categorized by data split.

《用于预测多环芳烃（Polycyclic Aromatic Hydrocarbons, PAHs）及其自由基阳离子代谢物致突变活性的可解释无代码符合经济合作与发展组织（Organization for Economic Co-operation and Development, OECD）标准的机器学习模型》（已投稿）补充材料：本研究所需全部配套支撑文件已整合至单个.ZIP压缩归档内。该归档包含两个核心文件夹：'Outputs'与'WEKA File'，二者分别存储激活态代谢物与前致癌物的分子结构对应的.xyz文件。'WEKA File'文件夹中还提供了用于模型训练与测试的.arff格式文件，所有文件均按具体的数据划分方式进行组织。'PAHs-GFN2-Data.xlsx'文件包含本研究使用的核心数据集，涵盖所有提取得到的分子细节、SMILES（简化分子线性输入规范，Simplified Molecular Input Line Entry System）符号、数值ID、CDFT（共轭密度泛函理论，Conjugated Density Functional Theory）描述符以及生物学响应数据，所有数据均按数据划分方式完成分类。