Machine learning approaches for analyzing and enhancing molecular dynamics simulations

被引:183
作者
Wang, Yihang [1 ,2 ]
Ribeiro, Joao Marcelo Lamim [3 ]
Tiwary, Pratyush [2 ,4 ]
机构
[1] Univ Maryland, Biophys Program, College Pk, MD 20742 USA
[2] Univ Maryland, Inst Phys Sci & Technol, College Pk, MD 20742 USA
[3] Icahn Sch Med Mt Sinai, Dept Pharmacol Sci, One Gustave L Levy Pl,Box 1677, New York, NY 10029 USA
[4] Univ Maryland, Dept Chem & Biochem, College Pk, MD 20742 USA
关键词
FREE-ENERGY LANDSCAPES;
D O I
10.1016/j.sbi.2019.12.016
中图分类号
Q5 [生物化学]; Q7 [分子生物学];
学科分类号
071010 ; 081704 ;
摘要
Molecular dynamics (MD) has become a powerful tool for studying biophysical systems, due to increasing computational power and availability of software. Although MD has made many contributions to better understanding these complex biophysical systems, there remain methodological difficulties to be surmounted. First, how to make the deluge of data generated in running even a microsecond long MD simulation human comprehensible. Second, how to efficiently sample the underlying free energy surface and kinetics. In this short perspective, we summarize machine learning based ideas that are solving both of these limitations, with a focus on their key theoretical underpinnings and remaining challenges.
引用
收藏
页码:139 / 145
页数:7
相关论文
共 43 条
  • [1] [Anonymous], 2018, ARXIV180507601
  • [2] [Anonymous], 2016, ARXIV160508803CSSTAT
  • [3] [Anonymous], 2019, ARTIFICIAL INTELLIGE
  • [4] Gaussian Approximation Potentials: The Accuracy of Quantum Mechanics, without the Electrons
    Bartok, Albert P.
    Payne, Mike C.
    Kondor, Risi
    Csanyi, Gabor
    [J]. PHYSICAL REVIEW LETTERS, 2010, 104 (13)
  • [5] Generalized neural-network representation of high-dimensional potential-energy surfaces
    Behler, Joerg
    Parrinello, Michele
    [J]. PHYSICAL REVIEW LETTERS, 2007, 98 (14)
  • [6] Neural networks-based variationally enhanced sampling
    Bonati, Luigi
    Zhang, Yue-Yu
    Parrinello, Michele
    [J]. PROCEEDINGS OF THE NATIONAL ACADEMY OF SCIENCES OF THE UNITED STATES OF AMERICA, 2019, 116 (36) : 17641 - 17647
  • [7] Promoting transparency and reproducibility in enhanced molecular simulations
    Bonomi, Massimiliano
    Bussi, Giovanni
    Camilloni, Carlo
    Tribello, Gareth A.
    Banas, Pavel
    Barducci, Alessandro
    Bernetti, Mattia
    Bolhuis, Peter G.
    Bottaro, Sandro
    Branduardi, Davide
    Capelli, Riccardo
    Carloni, Paolo
    Ceriotti, Michele
    Cesari, Andrea
    Chen, Haochuan
    Chen, Wei
    Colizzi, Francesco
    De, Sandip
    De La Pierre, Marco
    Donadio, Davide
    Drobot, Viktor
    Ensing, Bernd
    Ferguson, Andrew L.
    Filizola, Marta
    Fraser, James S.
    Fu, Haohao
    Gasparotto, Piero
    Gervasio, Francesco Luigi
    Giberti, Federico
    Gil-Ley, Alejandro
    Giorgino, Toni
    Heller, Gabriella T.
    Hocky, Glen M.
    Iannuzzi, Marcella
    Invernizzi, Michele
    Jelfs, Kim E.
    Jussupow, Alexander
    Kirilin, Evgeny
    Laio, Alessandro
    Limongelli, Vittorio
    Lindorff-Larsen, Kresten
    Lohr, Thomas
    Marinelli, Fabrizio
    Martin-Samos, Layla
    Masetti, Matteo
    Meyer, Ralf
    Michaelides, Angelos
    Molteni, Carla
    Morishita, Tetsuya
    Nava, Marco
    [J]. NATURE METHODS, 2019, 16 (08) : 670 - 673
  • [8] Machine Learning of Biomolecular Reaction Coordinates
    Brandt, Simon
    Sittel, Florian
    Ernst, Matthias
    Stock, Gerhard
    [J]. JOURNAL OF PHYSICAL CHEMISTRY LETTERS, 2018, 9 (09): : 2144 - 2150
  • [9] Simplifying the representation of complex free-energy landscapes using sketch-map
    Ceriotti, Michele
    Tribello, Gareth A.
    Parrinello, Michele
    [J]. PROCEEDINGS OF THE NATIONAL ACADEMY OF SCIENCES OF THE UNITED STATES OF AMERICA, 2011, 108 (32) : 13023 - 13028
  • [10] Capabilities and limitations of time-lagged autoencoders for slow mode discovery in dynamical systems
    Chen, Wei
    Sidky, Hythem
    Ferguson, Andrew L.
    [J]. JOURNAL OF CHEMICAL PHYSICS, 2019, 151 (06)