Grappa – A Machine Learned Molecular Mechanics Force Field
arxiv(2024)
摘要
Simulating large molecular systems over long timescales requires force fields
that are both accurate and efficient. In recent years, E(3) equivariant neural
networks have lifted the tension between computational efficiency and accuracy
of force fields, but they are still several orders of magnitude more expensive
than classical molecular mechanics (MM) force fields.
Here, we propose a novel machine learning architecture to predict MM
parameters from the molecular graph, employing a graph attentional neural
network and a transformer with symmetry-preserving positional encoding. The
resulting force field, Grappa, outperforms established and other
machine-learned MM force fields in terms of accuracy at the same computational
efficiency and can be used in existing Molecular Dynamics (MD) engines like
GROMACS and OpenMM. It predicts energies and forces of small molecules,
peptides, RNA and - showcasing its extensibility to uncharted regions of
chemical space - radicals at state-of-the-art MM accuracy. We demonstrate
Grappa's transferability to macromolecules in MD simulations, during which
large protein are kept stable and small proteins can fold. Our force field sets
the stage for biomolecular simulations close to chemical accuracy, but with the
same computational cost as established protein force fields.
更多查看译文
AI 理解论文
溯源树
样例
生成溯源树,研究论文发展脉络
Chat Paper
正在生成论文摘要