Ever wanted to run MD simulations of entire proteins in water with DFT accuracy?
Meet AMPv3-BMS25, the latest iteration of our AMP multiscale neural network potential by
@rinikerlab.bsky.social
Read more in the preprints:
doi.org/10.26434/che...
doi.org/10.26434/che...
We showcase its scalability and accuracy across various benchmarks (>23 μs!), modeling solvation free energies, protein structural features, and free-energy profiles of enzymatically catalyzed reactions.
Trained on the new BMS25 dataset, featuring over 1.5 million DFT calculations, it can perform these simulations on a single GPU. We demonstrate efficient scaling for systems comprising tens of thousands of protein atoms and hundreds of thousands of water molecules.
Published this week, our review articles on machine learning interatomic potentials in @chimiajournal.bsky.social with @rinikerlab.bsky.social : doi.org/10.2533/chim... we put special emphasis on ML/MM approaches that further reduce computational costs compared to DFT #compchem #ml
We have made the code, weights, and training dataset freely available for the community.
#ComputationalChemistry #MolecularDynamics #MachineLearning #StructuralBiology #DFT #AI #NNP #MLIP
