Release of ACEMD 4.0.0: Advancing Machine Learning for Molecular Dynamics Simulations

We are excited to announce the release of ACEMD 4.0.0, the latest version of our high-performance molecular dynamics engine. This release features the integration of Neural Network Potentials (NNP) with Molecular Mechanics (MM) workflows, a key advancement that combines the precision of machine learning with the efficiency of traditional methods, enabling hybrid simulations that offer the best of both worlds.

Why NNPs?

Neural Network Potentials bring the power of machine learning to molecular dynamics simulations, enabling a more accurate and efficient description of complex molecular systems. By training models on high-level quantum mechanical data, NNPs allow simulations to capture intricate atomic interactions with higher precision than traditional force fields, particularly in systems where standard MM methods struggle. This leads to enhanced accuracy in predicting molecular properties and behaviors, while maintaining the computational efficiency required for large-scale simulations. The hybrid approach of integrating NNPs with MM workflows in ACEMD 4.0.0 allows you to leverage both the speed of MM and the advanced accuracy of NNPs, providing a powerful tool for complex simulations. At Acellera Therapeutics we have already validated the advantages of using NNPs in a hybrid environment, such as relative binding free energy (RBFE) calculations.

What’s New in ACEMD 4.0.0:

• Simplified NNP/MM Simulations: Effortlessly integrate neural network potentials with molecular mechanics, enabling advanced hybrid simulations.
• Simplified Input Formats: Quickly configure simulations with the newly supported YAML or JSON file formats, streamlining the setup process.
• Enhanced Positional Restraints: New options provide greater precision and control over simulation constraints.
• Python API Integration: Run ACEMD directly from Python scripts, enabling tighter integration with custom workflows and tools.
• Backward Compatibility: Continue to use your ACEMD 3 inputs without modification, ensuring a smooth transition.

Discover the full documentation and start exploring ACEMD 4.0.0 today: https://software.acellera.com/acemd/

‍