Datasets for rapid assessment of virtually synthesizable chemical structures via support vector machine models

The " preprocessed " folder contains preprocessed compound datasets extracted from the ChEMBL 31 database for the 10 biological targets. These datasets can be used to generate reaction datasets consisting of active compounds with specific reaction templates. The eMolecules screening library compounds that match specific reaction templates (as reactants) were extracted and saved in the "emolecules" folder. A description of the reaction datasets can be found in "dataset_explanation.tsv". A detailed explanation of the usage of these datasets can be found in a GitHub repository at https://github.com/iwmspy/SVR-PK .