Reproduce “On the Selectivity of Generative Models in Structure-Based Drug Design” on OpenResearch
Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.
Universal app for Apple silicon and Intel. Requires macOS 11 or later.
About this paper
Ella Miray Rajaonson, Jungyoon Lee, William Chau, Alan Aspuru-Guzik, Benjamin Sanchez-Lengeling, Dominique Beaini, Kirill Neklyudov, Marta Skreta
The paper introduces SELECTBENCH to test whether structure-based drug design models favor an intended protein over competing proteins. It compares predicted binding scores across close-target pairs, safety-related panels, and synthetic pairs. For PARP1 versus PARP2, SpecLig’s normalized selectivity rate is 0.09 and TargetDiff’s is 0.07. These docking-based results do not establish experimental binding or clinical safety.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.