Electronic descriptors, molecular structures, and SISSO workflow for anthraquinone mutagenicity analysis

This dataset contains curated experimental mutagenicity classifications for 30 anthraquinones, standardized molecular structures in SDF format, computational results obtained from semiempirical electronic-structure calculations, and local electronic descriptors generated for the systematic analysis of molecular regions. It also includes Python scripts used for molecular preprocessing, region generation, descriptor extraction, preparation of SISSO input files, blockwise descriptor selection, threshold evaluation, and consolidation of the results.

The data were generated to investigate associations between local electronic structure and mutagenic activity and to support the reproducibility of the computational workflow. The repository includes experimental classifications, molecular structures, electronic-property data, descriptor tables, SISSO-related files, final ranking results, and instructions for reproducing the analyses