AMARO: All Heavy-Atom Transferable Neural Network Potentials of Protein Thermodynamics

被引：2

作者：

Mirarchi, Antonio ^{[1
]}

Pelaez, Raul P. ^{[1
]}

Simeon, Guillem ^{[1
]}

De Fabritiis, Gianni ^{[1
,2
,3
]}

机构：

[1] Univ Pompeu Fabra, Computat Sci Lab, Barcelona 08003, Spain

[2] Acellera Labs, Barcelona 08005, Spain

[3] Inst Catalana Recerca & Estudis Avancats ICREA, Barcelona 08010, Spain

来源：

JOURNAL OF CHEMICAL THEORY AND COMPUTATION | 2024年 / 20卷 / 22期

基金：

美国国家卫生研究院;

关键词：

MOLECULAR-DYNAMICS SIMULATIONS; COARSE-GRAINED MODEL; MARKOV STATE MODELS; FORCE-FIELD; KINETICS;

D O I：

10.1021/acs.jctc.4c01239

中图分类号：

O64 [物理化学（理论化学）、化学物理学];

学科分类号：

070304 ; 081704 ;

摘要：

All-atom molecular simulations offer detailed insights into macromolecular phenomena, but their substantial computational cost hinders the exploration of complex biological processes. We introduce Advanced Machine-learning Atomic Representation Omni-force-field (AMARO), a new neural network potential (NNP) that combines an O(3)-equivariant message-passing neural network architecture, TensorNet, with a coarse-graining map that excludes hydrogen atoms. AMARO demonstrates the feasibility of training coarser NNP, without prior energy terms, to run stable protein dynamics with scalability and generalization capabilities.

引用

页码：9871 / 9878

页数：8

共 50 条

[41] Embedded Atom Neural Network Potentials: Efficient and Accurate Machine Learning with a Physically Inspired Representation
Zhang, Yaolong
Hu, Ce
Jiang, Bin
JOURNAL OF PHYSICAL CHEMISTRY LETTERS, 2019, 10 (17): : 4962 - 4967
[42] De novo protein structure determination by heavy-atom soaking in lipidic cubic phase and SIKHS phasing using serial synchrotron crystallography
Botha, S.
Baitan, D.
Jungnickel, K. E. J.
Oberthuer, D.
Schmidt, C.
Stern, S.
Wiedorn, M. O.
Perbandt, M.
Chapman, H. N.
Betzel, C.
IUCRJ, 2018, 5 : 524 - 530
[43] CRYSTAL-GROWTH OF THE MEMBRANE-PROTEIN BACTERIORHODOPSIN - HEAVY-ATOM LABELING, AND LIQUID-LIQUID-DIFFUSION PROFILES IN MICROGRAVITY EXPERIMENTS
WAGNER, G
LINHARDT, R
PROCEEDINGS OF THE VIITH EUROPEAN SYMPOSIUM ON MATERIALS AND FLUID SCIENCES IN MICROGRAVITY, 1989, 295 : 35 - 39
[44] Pair-distribution-function guided optimization of fingerprints for atom-centered neural network potentials
Li, Lei
Li, Hao
Seymour, Ieuan D.
Koziol, Lucas
Henkelman, Graeme
JOURNAL OF CHEMICAL PHYSICS, 2020, 152 (22): : 224102
[45] Construction of high-dimensional neural network potentials using environment-dependent atom pairs
Jose, K. V. Jovan
Artrith, Nongnuch
Behler, Joerg
JOURNAL OF CHEMICAL PHYSICS, 2012, 136 (19):
[46] MULTIPLE HEAVY-ATOM SITES IN PROTEIN CRYSTALS HAVING CENTROSYMMETRIC PROJECTIONS - INTERPRETATION OF VECTOR MAPS .2. CORRELATION OF SITES IN DIFFERENT ADDUCTS
BUNN, CW
ACTA CRYSTALLOGRAPHICA SECTION B-STRUCTURAL CRYSTALLOGRAPHY AND CRYSTAL CHEMISTRY, 1971, B 27 (SEP15): : 1783 - &
[47] NOTE ON CORRELATION OF HEAVY-ATOM POSITIONS IN DIFFERENT ISOMORPHOUS PROTEIN CRYSTALS USING BOTH ISOMORPHOUS-REPLACEMENT AND ANOMALOUS-DISPERSION DATA
PARTHASARATHY, S
ACTA CRYSTALLOGRAPHICA SECTION A-CRYSTAL PHYSICS DIFFRACTION THEORETICAL AND GENERAL CRYSTALLOGRAPHY, 1971, A 27 (JAN1) : 45 - +
[48] Thermodynamics of an all-atom off-lattice model of the fragment B of Staphylococcal protein A:: Implication for the origin of the cooperativity of protein folding
Zhou, YQ
Linhananta, A
JOURNAL OF PHYSICAL CHEMISTRY B, 2002, 106 (06): : 1481 - 1485
[49] Enhancing Protein-Ligand Binding Affinity Predictions Using Neural Network Potentials
Zariquiey, Francesc Sabanes
Galvelis, Raimondas
Gallicchio, Emilio
Chodera, John D.
Markland, Thomas E.
De Fabritiis, Gianni
JOURNAL OF CHEMICAL INFORMATION AND MODELING, 2024, 64 (05) : 1481 - 1485
[50] Automatic Feature Selection for Atom-Centered Neural Network Potentials Using a Gradient Boosting Decision Algorithm
Li, Renzhe
Wang, Jiaqi
Singh, Akksay
Li, Bai
Song, Zichen
Zhou, Chuan
Li, Lei
JOURNAL OF CHEMICAL THEORY AND COMPUTATION, 2024, 20 (23) : 10564 - 10573

← 1 2 3 4 5 →