Reproduce “Empirical Bayes Flow Matching for Continuous Cryo-EM Heterogeneity” 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
Daniel Aibinder, Azmi Haider, Dan Rosenbaum
EB-FM learns a distribution of protein backbone conformations from cryo-EM projections by guiding candidate structures toward image consistency and training on persistent per-image estimates. On simulated data with known poses, it achieves an EMD-RMSD of 3.19 Å for protein 7r79 C, versus 4.76 Å for cryoDRGN-guided-AF3. The tests use backbone atoms only and EB-FM underestimates motion across all four proteins.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.