# Artificial Intelligence for Direct Prediction of Molecular Dynamics Across Chemical Space

arXiv: https://arxiv.org/abs/2505.16301 · HTML: https://mlsanity.com/paper/2505.16301/

## Key facts

- Cited by 2 papers in the map (data through 14 Oct 2026).
- Builds on 13 of its 53 references that are in the map.
- First posted on arXiv 22 May 2025.

## Builds on (most cited first, 13 of 13)

- [E(3)-Equivariant Graph Neural Networks for Data-Efficient and Accurate Interatomic Potentials](https://mlsanity.com/paper/2101.03164/) (2021)
- [SE(3)-Transformers: 3D Roto-Translation Equivariant Attention Networks](https://mlsanity.com/paper/2006.10503/) (2020)
- [Equivariant message passing for the prediction of tensorial properties and molecular spectra](https://mlsanity.com/paper/2102.03150/) (2021)
- [XCiT: Cross-Covariance Image Transformers](https://mlsanity.com/paper/2106.09681/) (2021)
- [Learning Local Equivariant Representations for Large-Scale Atomistic Dynamics](https://mlsanity.com/paper/2204.05249/) (2022)
- [Machine Learning Force Fields](https://mlsanity.com/paper/2010.07067/) (2020)
- [TorchMD-NET: Equivariant Transformers for Neural Network based Molecular Potentials](https://mlsanity.com/paper/2202.02541/) (2022)
- [Less is more: sampling chemical space with active learning](https://mlsanity.com/paper/1801.09319/) (2018)
- [Equivariant Transformer Networks](https://mlsanity.com/paper/1901.11399/) (2019)
- [A Comprehensive Survey of Transformers for Computer Vision](https://mlsanity.com/paper/2211.06004/) (2022)
- [MLatom 3: Platform for machine learning-enhanced computational chemistry simulations and workflows](https://mlsanity.com/paper/2310.20155/) (2023)
- [Learning Large-Time-Step Molecular Dynamics with Graph Neural Networks](https://mlsanity.com/paper/2111.15176/) (2021)
- [SE3ET: SE(3)-Equivariant Transformer for Low-Overlap Point Cloud Registration](https://mlsanity.com/paper/2407.16823/) (2024)

## Built on by (most cited first, 2 of 2)

- [FlashMD: long-stride, universal prediction of molecular dynamics](https://mlsanity.com/paper/2505.19350/) (2025)
- [Learning the action for long-time-step simulations of molecular dynamics](https://mlsanity.com/paper/2508.01068/) (2025)

## Abstract (arXiv)

Molecular dynamics (MD) is a powerful tool for exploring the behavior of atomistic systems, but its reliance on sequential numerical integration limits simulation efficiency. We present a novel neural network architecture, MDtrajNet, and a pre-trained foundational model, MDtrajNet-1, that directly generates MD trajectories across chemical space, bypassing force calculations and integration. MDtrajNet combines equivariant neural networks with a transformer-based architecture to achieve strong accuracy and transferability in predicting long-time trajectories. This approach accelerates simulations by up to two orders of magnitude and yields better accuracy than MD propagated with established machine-learning interatomic potentials trained on the same data. Remarkably, the errors of the trajectories generated by MDtrajNet-1 for various seen and even unseen small-sized molecular systems are close to those of the conventional ab initio MD. The current limitations of MDtrajNet-1 are attributed to the relatively small size of the chemical space in its training data; however, even for bigger, unseen systems, MDtrajNet-1 provides a good starting point for fine-tuning and obtaining system-specific models. The architecture's flexible design supports diverse application scenarios, including different statistical ensembles, boundary conditions, and interaction types. By overcoming the intrinsic speed barrier of conventional MD, MDtrajNet opens new frontiers in efficient and scalable atomistic simulations.