Reproduce “Learning consistent molecular mechanics force fields from first principles” 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
Berkay Günes, Leif Seute, Jigyasa Nigam, Frauke Gräter
Grappa-fullFF learns bonded and nonbonded parameters together for molecular mechanics, using quantum-mechanical energies and forces with electrostatic-potential and volume-based regularization. In 500 ns of explicit-solvent simulation of one capped alanine dipeptide in TIP3P water, its backbone-angle distribution had a Jensen–Shannon divergence of 0.0820 nats from the AMBER99SB-ILDN reference, versus 0.1596 for espaloma-0.3.2; this comparison does not establish broader transferability.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.