Reproduce “Generating and designing DNA with deep generative models” 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
Nathan Killoran, Leo J. Lee, Andrew Delong, David Duvenaud, Brendan J. Frey
The authors steer a DNA sequence generator by updating its latent input with gradients from a predictor of the desired property. For protein-binding probe design, models trained only on probes at or below the 40th percentile produced candidates that a separate oracle scored above the restricted training maximum, some at three times it. These are computational predictions, not wet-lab measurements.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.