4 papers
Emyx: Fast and efficient all-atom protein generation
Nicholas J. Williams, Ward Haddadin, Matteo P. Ferla +6
Computational enzyme design requires generating proteins that scaffold catalytic residues and ligands, a task that demands both geometric accuracy and structural diversity from the…
Global properties of the energy landscape: a testing and training arena for machine learned potentials
Vlad CÄrare, Fabian L. Thiemann, Joe Morrow +3
Machine learning interatomic potentials (MLIPs) have achieved remarkable accuracy on standard benchmarks, yet their ability to reproduce molecular kinetics -- critical for reaction…
Refining embeddings with fill-tuning: data-efficient generalised performance improvements for materials foundation models
Matthew P. Wilson, Edward O. Pyzer-Knapp, Nicolas Galichet +1
Pretrained foundation models learn embeddings that can be used for a wide range of downstream tasks. These embeddings optimise general performance, and if insufficiently accurate a…
Providing Machine Learning Potentials with High Quality Uncertainty Estimates
Zeynep Sumer, James L. McDonagh, Clyde Fare +4
Computational chemistry has come a long way over the course of several decades, enabling subatomic level calculations particularly with the development of Density Functional Theory…