Universally applicable and tunable graph-based coarse-graining for Machine learning force fields
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866908298423304192 |
|---|---|
| author | Brunken, Christoph Boyer, Sebastien Omar, Mustafa Maarand, Martin Peltre, Olivier Attias, Solal Diallo, Bakary N'tji Markina, Anastasia Othersen, Olaf Bent, Oliver |
| author_facet | Brunken, Christoph Boyer, Sebastien Omar, Mustafa Maarand, Martin Peltre, Olivier Attias, Solal Diallo, Bakary N'tji Markina, Anastasia Othersen, Olaf Bent, Oliver |
| contents | Coarse-grained (CG) force field methods for molecular systems are a crucial tool to simulate large biological macromolecules and are therefore essential for characterisations of biomolecular systems. While state-of-the-art deep learning (DL)-based models for all-atom force fields have improved immensely over recent years, we observe and analyse significant limitations of the currently available approaches for DL-based CG simulations. In this work, we present the first transferable DL-based CG force field approach (i.e., not specific to only one narrowly defined system type) applicable to a wide range of biosystems. To achieve this, our CG algorithm does not rely on hard-coded rules and is tuned to output coarse-grained systems optimised for minimal statistical noise in the ground truth CG forces, which results in significant improvement of model training. Our force field model is also the first CG variant that is based on the MACE architecture and is trained on a custom dataset created by a new approach based on the fragmentation of large biosystems covering protein, RNA and lipid chemistry. We demonstrate that our model can be applied in molecular dynamics simulations to obtain stable and qualitatively accurate trajectories for a variety of systems, while also discussing cases for which we observe limited reliability. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2504_01973 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Universally applicable and tunable graph-based coarse-graining for Machine learning force fields Brunken, Christoph Boyer, Sebastien Omar, Mustafa Maarand, Martin Peltre, Olivier Attias, Solal Diallo, Bakary N'tji Markina, Anastasia Othersen, Olaf Bent, Oliver Chemical Physics Artificial Intelligence Coarse-grained (CG) force field methods for molecular systems are a crucial tool to simulate large biological macromolecules and are therefore essential for characterisations of biomolecular systems. While state-of-the-art deep learning (DL)-based models for all-atom force fields have improved immensely over recent years, we observe and analyse significant limitations of the currently available approaches for DL-based CG simulations. In this work, we present the first transferable DL-based CG force field approach (i.e., not specific to only one narrowly defined system type) applicable to a wide range of biosystems. To achieve this, our CG algorithm does not rely on hard-coded rules and is tuned to output coarse-grained systems optimised for minimal statistical noise in the ground truth CG forces, which results in significant improvement of model training. Our force field model is also the first CG variant that is based on the MACE architecture and is trained on a custom dataset created by a new approach based on the fragmentation of large biosystems covering protein, RNA and lipid chemistry. We demonstrate that our model can be applied in molecular dynamics simulations to obtain stable and qualitatively accurate trajectories for a variety of systems, while also discussing cases for which we observe limited reliability. |
| title | Universally applicable and tunable graph-based coarse-graining for Machine learning force fields |
| topic | Chemical Physics Artificial Intelligence |
| url | https://arxiv.org/abs/2504.01973 |