skip to main content
DOE PAGES title logo U.S. Department of Energy
Office of Scientific and Technical Information

Title: A generalized deep learning approach for local structure identification in molecular simulations

Abstract

Identifying local structure in molecular simulations is of utmost importance. The most common existing approach to identify local structure is to calculate some geometrical quantity referred to as an order parameter. In simple cases order parameters are physically intuitive and trivial to develop (e.g., ion-pair distance), however in most cases, order parameter development becomes a much more difficult endeavor (e.g., crystal structure identification). Using ideas from computer vision, we adapt a specific type of neural network called a PointNet to identify local structural environments in molecular simulations. A primary challenge in applying machine learning techniques to simulation is selecting the appropriate input features. This challenge is system-specific and requires significant human input and intuition. In contrast, our approach is a generic framework that requires no system-specific feature engineering and operates on the raw output of the simulations, i.e., atomic positions. We demonstrate the method on crystal structure identification in Lennard-Jones (four different phases), water (eight different phases), and mesophase (six different phases) systems. The method achieves as high as 99.5% accuracy in crystal structure identification. The method is applicable to heterogeneous nucleation and it can even predict the crystal phases of atoms near external interfaces. We demonstrate the versatility ofmore » our approach by using our method to identify surface hydrophobicity based solely upon positions and orientations of surrounding water molecules. Our results suggest the approach will be broadly applicable to many types of local structure in simulations.« less

Authors:
ORCiD logo [1];  [2];  [1]; ORCiD logo [2]; ORCiD logo [1]
+ Show Author Affiliations
  1. Department of Chemical & Biomolecular Engineering, Clemson University, Clemson, USA
  2. Department of Electrical & Computer Engineering, Clemson University, Clemson, USA
Publication Date:
Research Org.:
Clemson Univ., SC (United States)
Sponsoring Org.:
USDOE Office of Science (SC), Basic Energy Sciences (BES)
OSTI Identifier:
1542673
Alternate Identifier(s):
OSTI ID: 1612289
Grant/Contract Number:  
SC0015448
Resource Type:
Published Article
Journal Name:
Chemical Science
Additional Journal Information:
Journal Name: Chemical Science Journal Volume: 10 Journal Issue: 32; Journal ID: ISSN 2041-6520
Publisher:
Royal Society of Chemistry
Country of Publication:
United Kingdom
Language:
English
Subject:
37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CHEMISTRY; Chemistry

Citation Formats

DeFever, Ryan S., Targonski, Colin, Hall, Steven W., Smith, Melissa C., and Sarupria, Sapna. A generalized deep learning approach for local structure identification in molecular simulations. United Kingdom: N. p., 2019. Web. doi:10.1039/C9SC02097G.
Copy to clipboard
DeFever, Ryan S., Targonski, Colin, Hall, Steven W., Smith, Melissa C., & Sarupria, Sapna. A generalized deep learning approach for local structure identification in molecular simulations. United Kingdom. doi:10.1039/C9SC02097G.
Copy to clipboard
DeFever, Ryan S., Targonski, Colin, Hall, Steven W., Smith, Melissa C., and Sarupria, Sapna. Thu . "A generalized deep learning approach for local structure identification in molecular simulations". United Kingdom. doi:10.1039/C9SC02097G.
Copy to clipboard
@article{osti_1542673,
title = {A generalized deep learning approach for local structure identification in molecular simulations},
author = {DeFever, Ryan S. and Targonski, Colin and Hall, Steven W. and Smith, Melissa C. and Sarupria, Sapna},
abstractNote = {Identifying local structure in molecular simulations is of utmost importance. The most common existing approach to identify local structure is to calculate some geometrical quantity referred to as an order parameter. In simple cases order parameters are physically intuitive and trivial to develop (e.g., ion-pair distance), however in most cases, order parameter development becomes a much more difficult endeavor (e.g., crystal structure identification). Using ideas from computer vision, we adapt a specific type of neural network called a PointNet to identify local structural environments in molecular simulations. A primary challenge in applying machine learning techniques to simulation is selecting the appropriate input features. This challenge is system-specific and requires significant human input and intuition. In contrast, our approach is a generic framework that requires no system-specific feature engineering and operates on the raw output of the simulations, i.e., atomic positions. We demonstrate the method on crystal structure identification in Lennard-Jones (four different phases), water (eight different phases), and mesophase (six different phases) systems. The method achieves as high as 99.5% accuracy in crystal structure identification. The method is applicable to heterogeneous nucleation and it can even predict the crystal phases of atoms near external interfaces. We demonstrate the versatility of our approach by using our method to identify surface hydrophobicity based solely upon positions and orientations of surrounding water molecules. Our results suggest the approach will be broadly applicable to many types of local structure in simulations.},
doi = {10.1039/C9SC02097G},
journal = {Chemical Science},
number = 32,
volume = 10,
place = {United Kingdom},
year = {2019},
month = {7}
}
Copy to clipboard
Journal Article:
Free Publicly Available Full Text
Publisher's Version of Record
DOI: 10.1039/C9SC02097G
Copyright Statement
Other availability
Search WorldCat Search WorldCat to find libraries that may hold this journal
Citation Metrics:
Cited by: 5 works
Citation information provided by
Web of Science
Save / Share:
Export Metadata
Save to My Library
You must Sign In or Create an Account in order to save documents to your library.

Works referenced in this record:

Accurate determination of crystal structures based on averaged local bond order parameters
journal, September 2008

  • Lechner, Wolfgang; Dellago, Christoph
  • The Journal of Chemical Physics, Vol. 129, Issue 11
  • DOI: 10.1063/1.2977970

Machine learning coarse grained models for water
journal, January 2019

Deformation-mechanism map for nanocrystalline metals by molecular-dynamics simulation
journal, December 2003

  • Yamakov, V.; Wolf, D.; Phillpot, S. R.
  • Nature Materials, Vol. 3, Issue 1
  • DOI: 10.1038/nmat1035

Why Is Gyroid More Difficult to Nucleate from Disordered Liquids than Lamellar and Hexagonal Mesophases?
journal, April 2018

  • Kumar, Abhinaw; Molinero, Valeria
  • The Journal of Physical Chemistry B, Vol. 122, Issue 17
  • DOI: 10.1021/acs.jpcb.8b02381

Deep learning
journal, May 2015

  • LeCun, Yann; Bengio, Yoshua; Hinton, Geoffrey
  • Nature, Vol. 521, Issue 7553
  • DOI: 10.1038/nature14539

Machine learning determination of atomic dynamics at grain boundaries
journal, October 2018

  • Sharp, Tristan A.; Thomas, Spencer L.; Cubuk, Ekin D.
  • Proceedings of the National Academy of Sciences, Vol. 115, Issue 43
  • DOI: 10.1073/pnas.1807176115

Seeding approach to crystal nucleation
journal, January 2016

  • Espinosa, Jorge R.; Vega, Carlos; Valeriani, Chantal
  • The Journal of Chemical Physics, Vol. 144, Issue 3
  • DOI: 10.1063/1.4939641

Quantifying Density Fluctuations in Volumes of All Shapes and Sizes Using Indirect Umbrella Sampling
journal, August 2011

  • Patel, Amish J.; Varilly, Patrick; Chandler, David
  • Journal of Statistical Physics, Vol. 145, Issue 2
  • DOI: 10.1007/s10955-011-0269-9

Molecular dynamics study of melting and freezing of small Lennard-Jones clusters
journal, September 1987

  • Honeycutt, J. Dana.; Andersen, Hans C.
  • The Journal of Physical Chemistry, Vol. 91, Issue 19, p. 4950-4963
  • DOI: 10.1021/j100303a014

Machine learning for crystal identification and discovery
journal, March 2018

  • Spellings, Matthew; Glotzer, Sharon C.
  • AIChE Journal, Vol. 64, Issue 6
  • DOI: 10.1002/aic.16157

Identifying Structural Flow Defects in Disordered Solids Using Machine-Learning Methods
journal, March 2015

Neural networks for local structure detection in polymorphic systems
journal, October 2013

  • Geiger, Philipp; Dellago, Christoph
  • The Journal of Chemical Physics, Vol. 139, Issue 16
  • DOI: 10.1063/1.4825111

Ice Nucleation Efficiency of Hydroxylated Organic Surfaces Is Controlled by Their Structural Fluctuations and Mismatch to Ice
journal, February 2017

  • Qiu, Yuqing; Odendahl, Nathan; Hudait, Arpa
  • Journal of the American Chemical Society, Vol. 139, Issue 8
  • DOI: 10.1021/jacs.6b12210

Predicting polymorphism in molecular crystals using orientational entropy
journal, September 2018

  • Piaggi, Pablo M.; Parrinello, Michele
  • Proceedings of the National Academy of Sciences, Vol. 115, Issue 41
  • DOI: 10.1073/pnas.1811056115

Characterizing Hydration Properties Based on the Orientational Structure of Interfacial Water Molecules
journal, January 2018

  • Shin, Sucheol; Willard, Adam P.
  • Journal of Chemical Theory and Computation, Vol. 14, Issue 2
  • DOI: 10.1021/acs.jctc.7b00898

Automated crystal characterization with a fast neighborhood graph analysis method
journal, January 2018

  • Reinhart, Wesley F.; Panagiotopoulos, Athanassios Z.
  • Soft Matter, Vol. 14, Issue 29
  • DOI: 10.1039/C8SM00960K

Collective variable discovery and enhanced sampling using autoencoders: Innovations in network architecture and error function design
journal, August 2018

  • Chen, Wei; Tan, Aik Rui; Ferguson, Andrew L.
  • The Journal of Chemical Physics, Vol. 149, Issue 7
  • DOI: 10.1063/1.5023804

The surface charge distribution affects the ice nucleating efficiency of silver iodide
journal, December 2016

  • Glatz, Brittany; Sarupria, Sapna
  • The Journal of Chemical Physics, Vol. 145, Issue 21
  • DOI: 10.1063/1.4966018

Machine-learning approach for local classification of crystalline structures in multiphase systems
journal, July 2017

Molecular dynamics simulations of ice growth from supercooled water
journal, November 2005

Machine learning for autonomous crystal structure identification
journal, January 2017

  • Reinhart, Wesley F.; Long, Andrew W.; Howard, Michael P.
  • Soft Matter, Vol. 13, Issue 27
  • DOI: 10.1039/C7SM00957G

NaCl nucleation from brine in seeded simulations: Sources of uncertainty in rate estimates
journal, June 2018

  • Zimmermann, Nils. E. R.; Vorselaars, Bart; Espinosa, Jorge R.
  • The Journal of Chemical Physics, Vol. 148, Issue 22
  • DOI: 10.1063/1.5024009

Direct calculation of ice homogeneous nucleation rate for a molecular model of water
journal, August 2015

  • Haji-Akbari, Amir; Debenedetti, Pablo G.
  • Proceedings of the National Academy of Sciences, Vol. 112, Issue 34
  • DOI: 10.1073/pnas.1509267112

Order parameters for the multistep crystallization of clathrate hydrates
journal, August 2011

  • Jacobson, Liam C.; Matsumoto, Masakazu; Molinero, Valeria
  • The Journal of Chemical Physics, Vol. 135, Issue 7
  • DOI: 10.1063/1.3613667

Self-Assembly of Mesophases from Nanoparticles
journal, September 2017

Automatic Method for Identifying Reaction Coordinates in Complex Systems †
journal, April 2005

  • Ma, Ao; Dinner, Aaron R.
  • The Journal of Physical Chemistry B, Vol. 109, Issue 14
  • DOI: 10.1021/jp045546c

Characterizing Solvent Density Fluctuations in Dynamical Observation Volumes
journal, January 2019

  • Jiang, Zhitong; Remsing, Richard C.; Rego, Nicholas B.
  • The Journal of Physical Chemistry B, Vol. 123, Issue 7
  • DOI: 10.1021/acs.jpcb.8b11423

Local order parameters for use in driving homogeneous ice nucleation with all-atom models of water
journal, November 2012

  • Reinhardt, Aleks; Doye, Jonathan P. K.; Noya, Eva G.
  • The Journal of Chemical Physics, Vol. 137, Issue 19
  • DOI: 10.1063/1.4766362

DeepIce: A Deep Neural Network Approach To Identify Ice and Water Molecules
journal, March 2019

  • Fulford, Maxwell; Salvalaglio, Matteo; Molteni, Carla
  • Journal of Chemical Information and Modeling, Vol. 59, Issue 5
  • DOI: 10.1021/acs.jcim.9b00005

Heterogeneous Ice Nucleation: Interplay of Surface Properties and Their Impact on Water Orientations
journal, October 2017

Crystal Nucleation in Liquids: Open Questions and Future Challenges in Molecular Dynamics Simulations
journal, May 2016

Nucleation mechanism of clathrate hydrates of water-soluble guest molecules
journal, November 2017

  • DeFever, Ryan S.; Sarupria, Sapna
  • The Journal of Chemical Physics, Vol. 147, Issue 20
  • DOI: 10.1063/1.4996132

Developing Local Order Parameters for Order–Disorder Transitions From Particles to Block Copolymers: Methodological Framework
journal, November 2018

Premelting, fluctuations, and coarse-graining of water-ice interfaces
journal, November 2014

  • Limmer, David T.; Chandler, David
  • The Journal of Chemical Physics, Vol. 141, Issue 18
  • DOI: 10.1063/1.4895399

Reweighted autoencoded variational Bayes for enhanced sampling (RAVE)
journal, August 2018

  • Ribeiro, João Marcelo Lamim; Bravo, Pablo; Wang, Yihang
  • The Journal of Chemical Physics, Vol. 149, Issue 7
  • DOI: 10.1063/1.5025487

Nonlinear dimensionality reduction in molecular simulation: The diffusion map approach
journal, June 2011

  • Ferguson, Andrew L.; Panagiotopoulos, Athanassios Z.; Kevrekidis, Ioannis G.
  • Chemical Physics Letters, Vol. 509, Issue 1-3
  • DOI: 10.1016/j.cplett.2011.04.066

Kinetic Pathways of Ion Pair Dissociation in Water
journal, May 1999

  • Geissler, Phillip L.; Dellago, Christoph; Chandler, David
  • The Journal of Physical Chemistry B, Vol. 103, Issue 18
  • DOI: 10.1021/jp984837g

Bond-orientational order in liquids and glasses
journal, July 1983

  • Steinhardt, Paul J.; Nelson, David R.; Ronchetti, Marco
  • Physical Review B, Vol. 28, Issue 2
  • DOI: 10.1103/PhysRevB.28.784

Hydrophobicity of proteins and nanostructured solutes is governed by topographical and chemical context
journal, November 2017

  • Xi, Erte; Venkateshwaran, Vasudevan; Li, Lijuan
  • Proceedings of the National Academy of Sciences, Vol. 114, Issue 51
  • DOI: 10.1073/pnas.1700092114

A simple method for displaying the hydropathic character of a protein
journal, May 1982

Robust structural identification via polyhedral template matching
journal, May 2016

  • Larsen, Peter Mahler; Schmidt, Søren; Schiøtz, Jakob
  • Modelling and Simulation in Materials Science and Engineering, Vol. 24, Issue 5
  • DOI: 10.1088/0965-0393/24/5/055007

A Simple Atomic-Level Hydrophobicity Scale Reveals Protein Interfacial Structure
journal, January 2014

Efficient Method To Characterize the Context-Dependent Hydrophobicity of Proteins
journal, January 2014

  • Patel, Amish J.; Garde, Shekhar
  • The Journal of Physical Chemistry B, Vol. 118, Issue 6
  • DOI: 10.1021/jp4081977

Pre-critical fluctuations and what they disclose about heterogeneous crystal nucleation
journal, December 2017

Generalized Neural-Network Representation of High-Dimensional Potential-Energy Surfaces
journal, April 2007

Role of stacking disorder in ice nucleation
journal, November 2017

  • Lupi, Laura; Hudait, Arpa; Peters, Baron
  • Nature, Vol. 551, Issue 7679
  • DOI: 10.1038/nature24279

Structure and Dynamics of the Quasi-Liquid Layer at the Surface of Ice from Molecular Simulations
journal, September 2018

  • Kling, Tanja; Kling, Felix; Donadio, Davide
  • The Journal of Physical Chemistry C, Vol. 122, Issue 43
  • DOI: 10.1021/acs.jpcc.8b07724

Reaction coordinates for the crystal nucleation of colloidal suspensions extracted from the reweighted path ensemble
journal, October 2011

  • Lechner, Wolfgang; Dellago, Christoph; Bolhuis, Peter G.
  • The Journal of Chemical Physics, Vol. 135, Issue 15
  • DOI: 10.1063/1.3651367

Adsorption of amino acids on graphene: assessment of current force fields
journal, January 2019

  • Dasetty, Siva; Barrows, John K.; Sarupria, Sapna
  • Soft Matter, Vol. 15, Issue 11
  • DOI: 10.1039/C8SM02621A

Neural network potential-energy surfaces in chemistry: a tool for large-scale simulations
journal, January 2011

  • Behler, Jörg
  • Physical Chemistry Chemical Physics, Vol. 13, Issue 40
  • DOI: 10.1039/c1cp21668f

Ice nucleation on carbon surface supports the classical theory for heterogeneous nucleation
journal, May 2015

Machine learning of accurate energy-conserving molecular force fields
journal, May 2017

  • Chmiela, Stefan; Tkatchenko, Alexandre; Sauceda, Huziel E.
  • Science Advances, Vol. 3, Issue 5
  • DOI: 10.1126/sciadv.1603015

Geometric diffusions as a tool for harmonic analysis and structure definition of data: Diffusion maps
journal, May 2005

  • Coifman, R. R.; Lafon, S.; Lee, A. B.
  • Proceedings of the National Academy of Sciences, Vol. 102, Issue 21
  • DOI: 10.1073/pnas.0500334102

Applications of local crystal structure measures in experiment and simulation
journal, February 2006

Mapping hydrophobicity at the nanoscale: Applications to heterogeneous surfaces and proteins
journal, January 2010

  • Acharya, Hari; Vembanur, Srivathsan; Jamadagni, Sumanth N.
  • Faraday Discussions, Vol. 146
  • DOI: 10.1039/b927019a

Two-component order parameter for quantifying clathrate hydrate nucleation and growth
journal, April 2014

  • Barnes, Brian C.; Beckham, Gregg T.; Wu, David T.
  • The Journal of Chemical Physics, Vol. 140, Issue 16
  • DOI: 10.1063/1.4871898

Self-Assembly of Colloidal Nanocrystals: From Intricate Structures to Functional Materials
journal, August 2016

Molecular Dynamics Study of Carbon Dioxide Hydrate Dissociation
journal, June 2011

  • Sarupria, Sapna; Debenedetti, Pablo G.
  • The Journal of Physical Chemistry A, Vol. 115, Issue 23
  • DOI: 10.1021/jp110868t

Identification of Clathrate Hydrates, Hexagonal Ice, Cubic Ice, and Liquid Water in Simulations: the CHILL+ Algorithm
journal, November 2014

  • Nguyen, Andrew H.; Molinero, Valeria
  • The Journal of Physical Chemistry B, Vol. 119, Issue 29
  • DOI: 10.1021/jp510289t

Self-assembly and applications of anisotropic nanomaterials: A review
journal, February 2015

    [ × clear filter / sort ]

    Similar Records in DOE PAGES and OSTI.GOV collections: