|
|
Physical information-enhanced graph neural network for predicting phase separation |
Yaqiang Zhang(张亚强)1, Xuwen Wang(王煦文)1, Yanan Wang(王雅楠)1, and Wen Zheng(郑文)1,2,3,† |
1 Institute of Public-Safety and Big Data, College of Computer Science and Technology (College of Data Science), Taiyuan University of Technology, Jinzhong 030600, China; 2 Shanxi Engineering Research Centre for Intelligent Data Assisted Treatment, Changzhi Medical College, Changzhi 046000, China; 3 Innovation Academy for Microsatellites of Chinese Academy of Sciences, Shanghai 200050, China |
|
|
Abstract Although phase separation is a ubiquitous phenomenon, the interactions between multiple components make it difficult to accurately model and predict. In recent years, machine learning has been widely used in physics simulations. Here, we present a physical information-enhanced graph neural network (PIENet) to simulate and predict the evolution of phase separation. The accuracy of our model in predicting particle positions is improved by 40.3% and 51.77% compared with CNN and SVM respectively. Moreover, we design an order parameter based on local density to measure the evolution of phase separation and analyze the systematic changes with different repulsion coefficients and different Schmidt numbers. The results demonstrate that our model can achieve long-term accurate predictions of order parameters without requiring complex handcrafted features. These results prove that graph neural networks can become new tools and methods for predicting the structure and properties of complex physical systems.
|
Received: 03 March 2024
Revised: 08 April 2024
Accepted manuscript online: 25 April 2024
|
PACS:
|
07.05.Mh
|
(Neural networks, fuzzy logic, artificial intelligence)
|
|
64.75.Gh
|
(Phase separation and segregation in model systems (hard spheres, Lennard-Jones, etc.))
|
|
83.10.Rs
|
(Computer simulation of molecular and particle dynamics)
|
|
87.64.Aa
|
(Computer simulation)
|
|
Fund: Project supported by the National Natural Science Foundation of China (Grant No. 11702289) and the Key Core Technology and Generic Technology Research and Development Project of Shanxi Province, China (Grant No. 2020XXX013). |
Corresponding Authors:
Wen Zheng
E-mail: zhengwen@tyut.edu.cn
|
Cite this article:
Yaqiang Zhang(张亚强), Xuwen Wang(王煦文), Yanan Wang(王雅楠), and Wen Zheng(郑文) Physical information-enhanced graph neural network for predicting phase separation 2024 Chin. Phys. B 33 070702
|
[1] Fletcher P D, Howe A M and Robinson B H 1987 Journal of the Chemical Society, Faraday Transactions 1: Physical Chemistry in Condensed Phases 83 985 [2] Buttinoni I, Bialké J, Kümmel F, Löwen H, Bechinger C and Speck T 2013 Phys. Rev. Lett. 110 238301 [3] Tung C, Harder J, Valeriani C and Cacciuto A 2016 Soft Matter 12 555 [4] Richard D, Löwen H and Speck T 2016 Soft Matter 12 5257 [5] Baumgart T, Hammond A T, Sengupta P, Hess S T, Holowka D A, Baird B A and Webb W W 2007 Proc. Nat. Acad. Sci. USA 104 3165 [6] Chong P A and Forman-Kay J D 2016 Current Opinion in Structural Biology 41 180 [7] Zola R S, Evangelista L, Yang Y C and Yang D K 2013 Phys. Rev. Lett. 110 057801 [8] Bažec M and Žumer S 2006 Phys. Rev. E 73 021703 [9] Jeon Y, Jamil M, Lee H D and Rhee J 2008 Pramana 71 559 [10] Connell S D and Smith D A 2006 Molecular Membrane Biology 23 17 [11] Girelli A, Rahmann H, Begam N, Ragulskaya A, Reiser M, Chandran S, Westermeier F, Sprung M, Zhang F, Gutt C, et al. 2021 Phys. Rev. Lett. 126 138004 [12] Novik K E and Coveney P V 2000 Phys. Rev. E 61 435 [13] Shimizu R and Tanaka H 2015 Nat. Commun. 6 7407 [14] Das K and Das S K 2023 Phys. Rev. E 107 044116 [15] Gidituri H, Akella V, Vedantam S and Panchagnula MV 2019 J. Chem. Phys. 150 234903 [16] Laradji M, Mouritsen O G, Toxvaerd S and Zuckermann M J 1994 Phys. Rev. E 50 1243 [17] Ahmad S, Das S K and Puri S 2012 Phys. Rev. E 85 031140 [18] Perlmutter J D and Sachs J N 2011 Biophys. J. 100 491a [19] Warren P B 1998 Current Opinion in Colloid & Interface Science 3 620 [20] Cai H, Vernon R M and Forman-Kay J D 2022 Biomolecules 12 1131 [21] Zhang P and Chern G W 2021 Phys. Rev. Lett. 127 146401 [22] Lopez C A, Vesselinov V V, Gnanakaran S and Alexandrov B S 2019 J. Chem. Theory Comput. 15 6343 [23] Chang Y C, Chen Y J, Chen P Y, Chen Y C, Maqbool F, Ho T Y and Chiang Y Y 2023 ACS Appl. Mater. Interfaces 15 12473 [24] Mandal R, Casert C and Sollich P 2022 Nat. Commun. 13 4424 [25] Chu X, Sun T, Li Q, Xu Y, Zhang Z, Lai L and Pei J 2022 BMC Bioinformatics 23 72 [26] Noe F, Tkatchenko A, Müller K R and Clementi C 2020 Annu. Rev. Phys. Chem. 71 361 [27] Zhang K, Li X, Jin Y and Jiang Y 2022 Soft Matter 18 6270 [28] Boattini E, Smallenburg F and Filion L 2021 Phys. Rev. Lett. 127 088007 [29] Mokhtar F, Kansal R, Diaz D, Duarte J M, Pata J, Pierini M and Vli mant J R 2021 arXiv:2111.12840v1 [physics.data-an] [30] Shen Z A, Luo T, Zhou Y K, Yu H and Du P F 2021 Briefings in Bioinformatics 22 bbab051 [31] Ha S and Jeong H 2021 Sci. Rep. 11 12804 [32] Wang R, Fang F, Cui J and Zheng W 2022 Sci. Rep. 12 500 [33] Zhao H, Wang R, Zhao C and Zheng W 2023 Chin. Phys. B 32 056402 [34] Liu Y, Wang R, Zhao C and Zheng W 2022 Chin. Phys. B 31 116401 [35] Sanchez-Gonzalez A, Godwin J, Pfaff T, Ying R, Leskovec J and Battaglia P 2020 Proceedings of the 37th International Conference on Machine Learning, July 13-18, 2020, Virtual, p. 8459 [36] Bapst V, Keck T, Grabska-Barwińska A, Donner C, Cubuk E D, Schoenholz S S, Obika A, Nelson A W, Back T, Hassabis D, et al. 2020 Nat. Phys. 16 448 [37] Shiba H, Hanai M, Suzumura T and Shimokawabe T 2023 J. Chem. Phys. 158 084503 [38] Groot R D and Warren P B 1997 J. Chem. Phys. 107 4423 [39] Visser D, Hoefsloot H C and Iedema P D 2006 J. Comput. Phys. 214 491 [40] Zhao Y, Liu H, Lu Z Y and Sun C C 2008 Chin. J. Chem. Phys. 21 451 |
No Suggested Reading articles found! |
|
|
Viewed |
|
|
|
Full text
|
|
|
|
|
Abstract
|
|
|
|
|
Cited |
|
|
|
|
Altmetric
|
blogs
Facebook pages
Wikipedia page
Google+ users
|
Online attention
Altmetric calculates a score based on the online attention an article receives. Each coloured thread in the circle represents a different type of online attention. The number in the centre is the Altmetric score. Social media and mainstream news media are the main sources that calculate the score. Reference managers such as Mendeley are also tracked but do not contribute to the score. Older articles often score higher because they have had more time to get noticed. To account for this, Altmetric has included the context data for other articles of a similar age.
View more on Altmetrics
|
|
|