Generative Active Learning for Molecular Design with REINVENT: Balancing Binding Affinity and Synthetic Accessibility

This dataset is used to generate all results and figures reported in the manuscript: "Generative Active Learning for Molecular Design with REINVENT: Balancing Binding Affinity and Synthetic Accessibility" Below are descriptions of individual files: GAL_AffSynth_Analysis.ipynb: Jupyter notebook for all analyses performed in the paper 3CLPro data: 3clpro_cazp_GAL_compounds.csv 3clpro_m1_GAL_compounds.csv 3clpro_nosa_GAL_compounds.csv 3clpro_ra_GAL_compounds.csv 3clpro_ram1_GAL_compounds.csv TNKS2 data: tnks2_cazp_GAL_compounds.csv tnks2_m1_GAL_compounds.csv tnks2_nosa_GAL_compounds.csv tnks2_ra_GAL_compounds.csv tnks2_ram1_GAL_compounds.csv