Inherent structural descriptors via machine learning

Telari, Emanuele; Tinti, Antonio; Settem, Manoj; Guardiani, Carlo; Kunche, Lakshmi Kumar; REES, MORGAN; HODDINOTT, HENRY; Dearg, Malcolm; Issendorff, Bernd von; Held, Georg; A, Thomas J; Palmer, Richard; Maragliano, Luca; Ferrando, Riccardo; Giacomello, Alberto

doi:10.1088/1361-6633/add95b

Journal article 418 views 54 downloads

Inherent structural descriptors via machine learning

Emanuele Telari

, 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)

Check full text

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...

Full description

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

first_indexed	2025-06-05T13:09:32Z
last_indexed	2025-06-06T07:03:44Z
id	cronfa69632
recordtype	SURis
fullrecord	<?xml version="1.0"?><rfc1807><datestamp>2025-06-05T14:13:42.5476751</datestamp><bib-version>v2</bib-version><id>69632</id><entry>2025-06-05</entry><title>Inherent structural descriptors via machine learning</title><swanseaauthors><author><sid>a6a581ddbeb110efced562c3c788261a</sid><firstname>MORGAN</firstname><surname>REES</surname><name>MORGAN REES</name><active>true</active><ethesisStudent>false</ethesisStudent></author><author><sid>cfc0d07768f530ee8ce43c8ad617e286</sid><firstname>HENRY</firstname><surname>HODDINOTT</surname><name>HENRY HODDINOTT</name><active>true</active><ethesisStudent>false</ethesisStudent></author><author><sid>6ae369618efc7424d9774377536ea519</sid><ORCID>0000-0001-8728-8083</ORCID><firstname>Richard</firstname><surname>Palmer</surname><name>Richard Palmer</name><active>true</active><ethesisStudent>false</ethesisStudent></author></swanseaauthors><date>2025-06-05</date><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.</abstract><type>Journal Article</type><journal>Reports on Progress in Physics</journal><volume>88</volume><journalNumber>6</journalNumber><paginationStart>068002</paginationStart><paginationEnd/><publisher>IOP Publishing</publisher><placeOfPublication/><isbnPrint/><isbnElectronic/><issnPrint>0034-4885</issnPrint><issnElectronic>1361-6633</issnElectronic><keywords>metal nanoclusters, machine learning, molecular dynamics, collective variables, free energy calculations</keywords><publishedDay>5</publishedDay><publishedMonth>6</publishedMonth><publishedYear>2025</publishedYear><publishedDate>2025-06-05</publishedDate><doi>10.1088/1361-6633/add95b</doi><url/><notes/><college>COLLEGE NANME</college><CollegeCode>COLLEGE CODE</CollegeCode><institution>Swansea University</institution><apcterm>Another institution paid the OA fee</apcterm><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.</funders><projectreference/><lastEdited>2025-06-05T14:13:42.5476751</lastEdited><Created>2025-06-05T14:00:46.5307205</Created><path><level id="1">Faculty of Science and Engineering</level><level id="2">School of Aerospace, Civil, Electrical, General and Mechanical Engineering - Mechanical Engineering</level></path><authors><author><firstname>Emanuele</firstname><surname>Telari</surname><orcid>0009-0009-3296-959X</orcid><order>1</order></author><author><firstname>Antonio</firstname><surname>Tinti</surname><orcid>0000-0002-6750-6503</orcid><order>2</order></author><author><firstname>Manoj</firstname><surname>Settem</surname><order>3</order></author><author><firstname>Carlo</firstname><surname>Guardiani</surname><orcid>0000-0002-8914-9260</orcid><order>4</order></author><author><firstname>Lakshmi Kumar</firstname><surname>Kunche</surname><order>5</order></author><author><firstname>MORGAN</firstname><surname>REES</surname><order>6</order></author><author><firstname>HENRY</firstname><surname>HODDINOTT</surname><order>7</order></author><author><firstname>Malcolm</firstname><surname>Dearg</surname><order>8</order></author><author><firstname>Bernd von</firstname><surname>Issendorff</surname><order>9</order></author><author><firstname>Georg</firstname><surname>Held</surname><order>10</order></author><author><firstname>Thomas J A</firstname><surname>Slater</surname><orcid>0000-0003-0372-1551</orcid><order>11</order></author><author><firstname>Richard</firstname><surname>Palmer</surname><orcid>0000-0001-8728-8083</orcid><order>12</order></author><author><firstname>Luca</firstname><surname>Maragliano</surname><orcid>0000-0002-5705-6967</orcid><order>13</order></author><author><firstname>Riccardo</firstname><surname>Ferrando</surname><orcid>0000-0003-2750-9061</orcid><order>14</order></author><author><firstname>Alberto</firstname><surname>Giacomello</surname><orcid>0000-0003-2735-6982</orcid><order>15</order></author></authors><documents><document><filename>69632__34397__a6945c2649a8464ba834c05921f3c4a8.pdf</filename><originalFilename>pdf.pdf</originalFilename><uploaded>2025-06-05T14:00:46.5300726</uploaded><type>Output</type><contentLength>2981901</contentLength><contentType>application/pdf</contentType><version>Version of Record</version><cronfaStatus>true</cronfaStatus><documentNotes>© 2025 The Author(s). Original Content from this work may be used under the terms of the Creative Commons Attribution 4.0 licence.</documentNotes><copyrightCorrect>true</copyrightCorrect><language>eng</language><licence>https://creativecommons.org/licenses/by/4.0/</licence></document></documents><OutputDurs/></rfc1807>
spelling	2025-06-05T14:13:42.5476751 v2 69632 2025-06-05 Inherent structural descriptors via machine learning a6a581ddbeb110efced562c3c788261a MORGAN REES MORGAN REES true false cfc0d07768f530ee8ce43c8ad617e286 HENRY HODDINOTT HENRY HODDINOTT true false 6ae369618efc7424d9774377536ea519 0000-0001-8728-8083 Richard Palmer Richard Palmer true false 2025-06-05 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. Journal Article Reports on Progress in Physics 88 6 068002 IOP Publishing 0034-4885 1361-6633 metal nanoclusters, machine learning, molecular dynamics, collective variables, free energy calculations 5 6 2025 2025-06-05 10.1088/1361-6633/add95b COLLEGE NANME COLLEGE CODE Swansea University Another institution paid the OA fee 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. 2025-06-05T14:13:42.5476751 2025-06-05T14:00:46.5307205 Faculty of Science and Engineering School of Aerospace, Civil, Electrical, General and Mechanical Engineering - Mechanical Engineering Emanuele Telari 0009-0009-3296-959X 1 Antonio Tinti 0000-0002-6750-6503 2 Manoj Settem 3 Carlo Guardiani 0000-0002-8914-9260 4 Lakshmi Kumar Kunche 5 MORGAN REES 6 HENRY HODDINOTT 7 Malcolm Dearg 8 Bernd von Issendorff 9 Georg Held 10 Thomas J A Slater 0000-0003-0372-1551 11 Richard Palmer 0000-0001-8728-8083 12 Luca Maragliano 0000-0002-5705-6967 13 Riccardo Ferrando 0000-0003-2750-9061 14 Alberto Giacomello 0000-0003-2735-6982 15 69632__34397__a6945c2649a8464ba834c05921f3c4a8.pdf pdf.pdf 2025-06-05T14:00:46.5300726 Output 2981901 application/pdf Version of Record true © 2025 The Author(s). Original Content from this work may be used under the terms of the Creative Commons Attribution 4.0 licence. true eng https://creativecommons.org/licenses/by/4.0/
title	Inherent structural descriptors via machine learning
spellingShingle	Inherent structural descriptors via machine learning MORGAN REES HENRY HODDINOTT Richard Palmer
title_short	Inherent structural descriptors via machine learning
title_full	Inherent structural descriptors via machine learning
title_fullStr	Inherent structural descriptors via machine learning
title_full_unstemmed	Inherent structural descriptors via machine learning
title_sort	Inherent structural descriptors via machine learning
author_id_str_mv	a6a581ddbeb110efced562c3c788261a cfc0d07768f530ee8ce43c8ad617e286 6ae369618efc7424d9774377536ea519
author_id_fullname_str_mv	a6a581ddbeb110efced562c3c788261a_*_MORGAN REES cfc0d07768f530ee8ce43c8ad617e286__HENRY HODDINOTT 6ae369618efc7424d9774377536ea519_**_Richard Palmer
author	MORGAN REES HENRY HODDINOTT Richard Palmer
author2	Emanuele Telari 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
format	Journal article
container_title	Reports on Progress in Physics
container_volume	88
container_issue	6
container_start_page	068002
publishDate	2025
institution	Swansea University
issn	0034-4885 1361-6633
doi_str_mv	10.1088/1361-6633/add95b
publisher	IOP Publishing
college_str	Faculty of Science and Engineering
hierarchytype
hierarchy_top_id	facultyofscienceandengineering
hierarchy_top_title	Faculty of Science and Engineering
hierarchy_parent_id	facultyofscienceandengineering
hierarchy_parent_title	Faculty of Science and Engineering
department_str	School of Aerospace, Civil, Electrical, General and Mechanical Engineering - Mechanical Engineering{{{_:::_}}}Faculty of Science and Engineering{{{_:::_}}}School of Aerospace, Civil, Electrical, General and Mechanical Engineering - Mechanical Engineering
document_store_str	1
active_str	0
description	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.
published_date	2025-06-05T05:21:03Z
_version_	1857710994893045760
score	11.096892

Inherent structural descriptors via machine learning

Similar Items