Journal article 316 views 43 downloads
Inherent structural descriptors via machine learning
,
Antonio Tinti ,
Manoj Settem,
Carlo Guardiani ,
Lakshmi Kumar Kunche,
MORGAN REES,
HENRY HODDINOTT,
Malcolm Dearg,
Bernd von Issendorff,
Georg Held,
Thomas J A Slater ,
Richard Palmer ,
Luca Maragliano ,
Riccardo Ferrando ,
Alberto Giacomello
Reports on Progress in Physics, Volume: 88, Issue: 6, Start page: 068002
Swansea University Authors:
MORGAN REES, HENRY HODDINOTT, Richard Palmer
-
PDF | Version of Record
© 2025 The Author(s). Original Content from this work may be used under the terms of the Creative Commons Attribution 4.0 licence.
Download (2.84MB)
DOI (Published version): 10.1088/1361-6633/add95b
Abstract
Finding proper collective variables for complex systems and processes is one of the most challenging tasks in simulations, which limits the interpretation of experimental and simulated data and the application of enhanced sampling techniques. Here, we propose a machine learning (ML) approach able to...
| Published in: | Reports on Progress in Physics |
|---|---|
| ISSN: | 0034-4885 1361-6633 |
| Published: |
IOP Publishing
2025
|
| Online Access: |
Check full text
|
| URI: | https://cronfa.swan.ac.uk/Record/cronfa69632 |
| Abstract: |
Finding proper collective variables for complex systems and processes is one of the most challenging tasks in simulations, which limits the interpretation of experimental and simulated data and the application of enhanced sampling techniques. Here, we propose a machine learning (ML) approach able to distill few, physically relevant variables by associating instantaneous configurations of the system to their corresponding inherent structures as defined in liquids theory. We apply this approach to the challenging case of structural transitions in nanoclusters, managing to characterize and explore the structural complexity of an experimentally relevant system constituted by 147 gold atoms. Our inherent-structure variables are shown to be effective at computing complex free-energy landscapes, transition rates, and at describing non-equilibrium melting and freezing processes. In addition, we illustrate the generality of this ML strategy by deploying it to understand conformational rearrangements of the bradykinin peptide, indicating its applicability to a vast range of systems, including liquids, glasses, and proteins. |
|---|---|
| Keywords: |
metal nanoclusters, machine learning, molecular dynamics, collective variables, free energy calculations |
| College: |
Faculty of Science and Engineering |
| Funders: |
We acknowledge financial support under the National Recovery and Resilience Plan (NRRP), Mission 4, Component 2, Investment 1.1, Call for tender No. 104 published on 2.2.2022 by the Italian Ministry of University and Research (MUR), funded by the European Union—NextGenerationEU- Project Title PINENUT—CUP D53D23002340006 - Grant Assignment Decree No. 957 adopted on 30/06/2023 by the Italian Ministry of University and Research (MUR). We thank Diamond Light Source for access to and support in use of the electron Physical Science Imaging Centre (Instrument E02, Proposal No.: MG28449), and gratefully acknowledge EPSRC Grant EP/V029797/2 for support of the electron microscopy. MR is grateful to EPSRC (via Swansea University) and Johnson Matthey for a PhD scholarship. HH is grateful to EPSRC (via the M2A CDT at Swansea University) and Diamond Light Source for an EngD scholarship. |
| Issue: |
6 |
| Start Page: |
068002 |