Reproduce “Automatic chemical design using a data-driven continuous representation of molecules” 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
Rafael Gómez-Bombarelli, Jennifer N. Wei, David Duvenaud, José Miguel Hernández-Lobato, Benjamín Sánchez-Lengeling, Dennis Sheberla, Jorge Aguilera-Iparraguirre, Timothy D. Hirzel and 2 more
The authors learn a continuous latent representation of molecules from SMILES strings and use a Gaussian-process surrogate to guide search. On ZINC drug-like molecules, optimizing proxy score 5 × QED − SAS from the bottom 10% produced higher objective percentiles than random Gaussian search or a genetic algorithm. This is a computational result, not experimental evidence of useful compounds.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.