Advancing Machine-Learned Interatomic Potentials for Proteins Simulation (MLP4ProSim): From Development to Application

Atomic simulation of the proteins is one of the main tools in understanding the functions, interactions, and dynamics of these vitally important biomolecules and plays a pivotal role in numerous cutting-edge researches such as elucidating the pathways of diseases, drug discovery and so on.

The limited accuracy of the classical force fields specially in approximating non-bonded interactions as well as in capturing kinetics of proteins has attracted attentions towards more advanced alternatives.

MLP4ProSim aims to develop next-generation transferable all-atom machine-learned potentials (MLPs) for computer simulation of proteins.

To assure the applicability of the developed potentials for more realistic and biologically relevant simulations, training of the MLPs is carried out using a critically assessed dataset containing both quantum-mechanically computed and experimentally determined reference data.

Enabling transferability of the developed MLPs and their applicability for the simulation of different proteins without requiring demanding retraining processes will be addressed by considering training sets containing proteins with diverse shapes, functions, and sizes.

The resulting MLPs, after being validated for reliability and transferability via extensive benchmark tests, will be implemented in LAMMPS as an extensively applied molecular dynamics simulation code to make it openly accessible to the community.

The successful completion of the project objectives is expected to have significant impacts in a broad range of scientific and technological areas.

In addition to the anticipated outstanding scientific impacts, this project encompasses comprehensive trainings planned for the experienced researcher Dr. Amin Alibakhshi in accordance with the objectives of the Horizon Europe Framework Programme.

This training aims to equip him with the essential skills and innovative capabilities required to emerge as a future leader at the forefront of academic excellence.