Findings demonstrate structural models enhance virtual screening for ligand classification, highlighting AI's role.
We provide the raw data to reproduce results presented in our manuscript "On the rise of AI technologies in virtual screening" submitted to JCIM. The data include: The raw data required to reproduce ligand classification by Boltz-2 in the ULVSH dataset (TXT) Structural models by Boltz-2 for all protein-ligand complexes in the ULVSH dataset (pymol sessions) The docking hit lists from the LSD dataset analyzed in this work. These lists were produced by extracting the top-1000 post-docking hits per target and mixing them with all experimentally confirmed actives including those that were ranked lower by docking (CSV)
No takes yet. Share an insight, caveat, or question.
Marco Cecchini (2025) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: