Training and sampled datasets and PDB files of synthesized compounds for "Benchmarking Generative Models for Hit Generation Across Multiple Targets"

This repository provides the datasets and structural files used in the study: "Benchmarking Generative Models for Hit Generation Across Multiple Targets" Included materials:- Molecular datasets employed for model training- Generated compounds produced by the generative models across all algorithms and training methods- PDB structures associated with docking studies of synthesized molecules- Supplementary CSV files containing SMILES strings and experimental data for each synthesized compound These materials are made available to support reproducibility, benchmarking, and downstream analysis of the reported results.