Data science techniques in biomolecular force field development

Article Chemistry, Physical

Deep Neural Network Model to Predict the Electrostatic Parameters in the Polarizable Classical Drude Oscillator Force Field

Anmol Kumar et al.

Summary: The Drude polarizable force field captures electronic polarization effects and is useful for simulating biomolecules and other molecules. Deep neural network models can accurately predict molecular dipole moments and polarizabilities.

JOURNAL OF CHEMICAL THEORY AND COMPUTATION (2022)

Add to Collection

Article Chemistry, Physical

Harnessing Deep Learning for Optimization of Lennard-Jones Parameters for the Polarizable Classical Drude Oscillator Force Field

Payal Chatterjee et al.

Summary: The outcomes of computational chemistry and biology research are greatly influenced by the choice of forcefield used in molecular simulations. This study develops a deep learning-based framework for optimizing van der Waals interactions in molecular simulations. The resulting LJ parameters (interactions between atoms) are validated for their performance in reproducing condensed phase thermodynamic properties and demonstrate improved accuracy in reproducing solvent and crystal properties.

JOURNAL OF CHEMICAL THEORY AND COMPUTATION (2022)

Add to Collection

Article Chemistry, Physical

BuRNN: Buffer Region Neural Network Approach for Polarizable-Embedding Neural Network/Molecular Mechanics Simulations

Bettina Lier et al.

Summary: Hybrid quantum mechanics/molecular mechanics (QM/MM) simulations have greatly advanced computational chemistry, but their partitioning of a system into different regions and treating them with different theories can cause artifacts at the interface. In this study, the buffer region neural network (BuRNN) is introduced as an alternative approach to minimize these artifacts by fully polarizing a buffer region. The interaction between the quantum mechanics (QM) and buffer regions is described by deep neural networks (NNs), resulting in high computational efficiency and quantum chemical accuracy.

JOURNAL OF PHYSICAL CHEMISTRY LETTERS (2022)

Add to Collection

Article Chemistry, Multidisciplinary

End-to-end differentiable construction of molecular mechanics force fields

Yuanqing Wang et al.

Summary: A new approach using graph neural networks is proposed to better perceive chemical environments and predict parameters, facilitating the construction, extension, and application of new force fields for arbitrary molecules with higher accuracy compared to traditional methods.

CHEMICAL SCIENCE (2022)

Add to Collection

Article Chemistry, Physical

Non-Covalent Interactions Atlas benchmark data sets 5: London dispersion in an extended chemical space

Jan Rezac

Summary: The Non-Covalent Interactions Atlas has expanded with two data sets to study benchmark interaction energies in complexes dominated by London dispersion. These high-quality data sets enable comprehensive testing of various non-covalent interactions in a wide range of chemical space.

PHYSICAL CHEMISTRY CHEMICAL PHYSICS (2022)

Add to Collection

Article Chemistry, Multidisciplinary

Validating the CHARMM36m protein force field with LJ-PME reveals altered hydrogen bonding dynamics under elevated pressures

You Xu et al.

Summary: The pressure-temperature phase diagram plays a crucial role in understanding the physics of biomolecules. The transferability of the CHARMM36m protein force field has been evaluated under different pressures, showing good agreement with experimental data and suitability for exploring protein dynamics. The study highlights the potential application of CHARMM36m in investigating protein behavior under elevated pressures.

COMMUNICATIONS CHEMISTRY (2021)

Add to Collection

Article Chemistry, Physical

Development of Range-Corrected Deep Learning Potentials for Fast, Accurate Quantum Mechanical/Molecular Mechanical Simulations of Chemical Reactions in Solution

Jinzhe Zeng et al.

Summary: The new deep potential-range correction (DPRc) machine learning potential and training procedure offer accurate QM/MM interaction corrections for efficient simulation of chemical reactions. Tested DPRc model can quickly and accurately reproduce free-energy profiles, showing promising potential for wide-ranging applications.

JOURNAL OF CHEMICAL THEORY AND COMPUTATION (2021)

Add to Collection

Review Chemistry, Multidisciplinary

Machine Learning Force Fields

Oliver T. Unke et al.

Summary: The use of machine learning in computational chemistry has led to significant advancements, particularly in the development of machine learning-based force fields to bridge the gap between accuracy and efficiency. The key concept is to learn the statistical relations between chemical structure and potential energy, without preconceived notions of fixed bonds. Challenges remain for the next generation of machine learning force fields.

CHEMICAL REVIEWS (2021)

Add to Collection

Article Multidisciplinary Sciences

Accurate model of liquid-liquid phase behavior of intrinsically disordered proteins from optimization of single-chain properties

Giulio Tesei et al.

Summary: Researchers have found that many intrinsically disordered proteins (IDPs) can participate in the formation of membraneless organelles in the cell. Investigating the phase behavior of IDPs through molecular simulations requires an accurate description of intramolecular and intermolecular interactions. By developing a coarse-grained model based on experimental data, the model accurately reproduces the conformational propensities of IDPs.

PROCEEDINGS OF THE NATIONAL ACADEMY OF SCIENCES OF THE UNITED STATES OF AMERICA (2021)

Add to Collection

Article Multidisciplinary Sciences

Quantum chemical benchmark databases of gold-standard dimer interaction energies

Alexander G. Donchev et al.

Summary: This paper introduces three benchmark collections of quantum mechanical data encompassing approximately 3,700 different types of noncovalent molecular interactions. These datasets can be useful for research in electronic structure theory and computational chemistry.

SCIENTIFIC DATA (2021)

Add to Collection

Article Biochemistry & Molecular Biology

Systematic design of biomolecular force fields

David van der Spoel

Summary: Force fields for the study of biomolecules have evolved predominantly in an organic manner over the past half century, with regular updates. Advancements in algorithms and computer technology have led to more robust predictions. However, there is still room for improvement, with efforts needed to develop novel force fields from scratch and standardize reference data and validation benchmarks.

CURRENT OPINION IN STRUCTURAL BIOLOGY (2021)

Add to Collection

Article Chemistry, Physical

TorchMD: A Deep Learning Framework for Molecular Simulations

Stefan Doerr et al.

Summary: TorchMD is a molecular simulation framework that utilizes both classical and machine learning potentials. It allows for various force computations and supports learning and simulating neural network potentials. Through validation, it has been proven to be a useful toolkit for molecular simulations with machine learning potentials.

JOURNAL OF CHEMICAL THEORY AND COMPUTATION (2021)

Add to Collection

Article Multidisciplinary Sciences

Machine learning-accelerated computational fluid dynamics

Dmitrii Kochkov et al.

Summary: Numerical simulation of fluids is crucial in modeling physical phenomena like weather and aerodynamics, but solving the Navier-Stokes equations at scale remains challenging due to computational costs. This study uses end-to-end deep learning to improve approximations in computational fluid dynamics for modeling two-dimensional turbulent flows, achieving computational speedups and stability while generalizing well to different scenarios. Leveraging machine learning and hardware accelerators, this approach showcases the potential to enhance simulations without sacrificing accuracy or generalization.

PROCEEDINGS OF THE NATIONAL ACADEMY OF SCIENCES OF THE UNITED STATES OF AMERICA (2021)

Add to Collection

Article Physics, Condensed Matter

Machine learning potentials for extended systems: a perspective

Joerg Behler et al.

Summary: In the past two and a half decades, machine learning potentials have evolved to become a broadly applicable tool for large-scale atomistic simulations, combining efficiency and accuracy. These models exploit the locality of atomic properties and can include long-range interactions, even for systems with non-local charge transfer. Challenges and limitations of current approaches are also discussed.

EUROPEAN PHYSICAL JOURNAL B (2021)

Add to Collection

Article Chemistry, Physical

Machine-Learning-Assisted Free Energy Simulation of Solution-Phase and Enzyme Reactions

Xiaoliang Pan et al.

Summary: This study presents a protocol for machine-learning-assisted free energy simulation of solution-phase and enzyme reactions at the ab initio quantum-mechanical/molecular-mechanical (ai-QM/MM) level. The developed MLP and Delta MLP exhibit promising accuracy in reproducing energy and forces, offering a cost-effective alternative for studying enzymatic reactions.

JOURNAL OF CHEMICAL THEORY AND COMPUTATION (2021)

Add to Collection

Article Multidisciplinary Sciences

Highly accurate protein structure prediction with AlphaFold

John Jumper et al.

Summary: Proteins are essential for life, and accurate prediction of their structures is a crucial research problem. Current experimental methods are time-consuming, highlighting the need for accurate computational approaches to address the gap in structural coverage. Despite recent progress, existing methods fall short of atomic accuracy in protein structure prediction.

NATURE (2021)

Add to Collection

Article Multidisciplinary Sciences

Differentiable molecular simulation can learn all the parameters in a coarse-grained force field for proteins

Joe G. Greener et al.

Summary: The study utilizes automatic differentiation to optimize force field parameters in molecular simulation, achieving real-time parameter adjustments while retaining interpretability and efficiency of the force field. By training a simple coarse-grained force field, the potential matches chemical knowledge and PDB data, demonstrating potential in protein design and model scoring applications.

PLOS ONE (2021)

Add to Collection

Article Computer Science, Interdisciplinary Applications

Biomolecular modeling thrives in the age of technology

Tamar Schlick et al.

Summary: The field of biomolecular modeling has thrived since the 1970s, thanks to rapid adaptation and tailoring of cutting-edge technology. Hardware and software advancements play a crucial role in driving the field forward. A bright future for modeling is indicated by the rapid adaptation and outreach, where theory, experimentation, and simulation are the three pillars needed to address future scientific and biomedical challenges.

NATURE COMPUTATIONAL SCIENCE (2021)

Add to Collection

Article Biochemistry & Molecular Biology

PED in 2021: a major update of the protein ensemble database for intrinsically disordered proteins

Tamas Lazar et al.

Summary: The Protein Ensemble Database (PED) has been significantly updated to version 4.0, with cutting-edge technology and an increased amount of data. The database now includes a renewed graphical interface, interactive feature viewer, and high-quality data guaranteed by a new submission process. Additionally, a new search engine allows users to build advanced queries and search all entry fields.

NUCLEIC ACIDS RESEARCH (2021)

Add to Collection

Article Computer Science, Interdisciplinary Applications

DP-GEN: A concurrent learning platform for the generation of reliable deep learning based potential energy models

Yuzhi Zhang et al.

COMPUTER PHYSICS COMMUNICATIONS (2020)

Add to Collection

Article Chemistry, Physical

Non-Covalent Interactions Atlas Benchmark Data Sets: Hydrogen Bonding

Jan Rezac

JOURNAL OF CHEMICAL THEORY AND COMPUTATION (2020)

Add to Collection

Article Multidisciplinary Sciences

The ANI-1ccx and ANI-1x data sets, coupled-cluster and density functional theory properties for molecules

Justin S. Smith et al.

SCIENTIFIC DATA (2020)

Add to Collection

Article Chemistry, Physical

Machine learning for collective variable discovery and enhanced sampling in biomolecular simulation

Hythem Sidky et al.

MOLECULAR PHYSICS (2020)

Add to Collection

Article Chemistry, Physical

Extending the Applicability of the ANI Deep Learning Molecular Potential to Sulfur and Halogens

Christian Devereux et al.

JOURNAL OF CHEMICAL THEORY AND COMPUTATION (2020)

Add to Collection

Article Physics, Multidisciplinary

Propagation of uncertainty in physicochemical data to force field predictions

Ahmet Yildirim et al.

PHYSICAL REVIEW RESEARCH (2020)

Add to Collection

Article Multidisciplinary Sciences

Boltzmann generators: Sampling equilibrium states of many-body systems with deep learning

Frank Noe et al.

SCIENCE (2019)

Add to Collection

Article Computer Science, Interdisciplinary Applications

DeePMD-kit: A deep learning package for many-body potential energy representation and molecular dynamics

Han Wang et al.

COMPUTER PHYSICS COMMUNICATIONS (2018)

Add to Collection

Article Biochemistry & Molecular Biology

New developments in force fields for biomolecular simulations

Paul S. Nerenberg et al.

CURRENT OPINION IN STRUCTURAL BIOLOGY (2018)

Add to Collection

Article Biochemistry & Molecular Biology

Force field development and simulations of intrinsically disordered proteins

Jing Huang et al.

CURRENT OPINION IN STRUCTURAL BIOLOGY (2018)

Add to Collection

Article Multidisciplinary Sciences

Developing a molecular dynamics force field for both folded and disordered protein states

Paul Robustelli et al.

PROCEEDINGS OF THE NATIONAL ACADEMY OF SCIENCES OF THE UNITED STATES OF AMERICA (2018)

Add to Collection

Article Multidisciplinary Sciences

Solving high-dimensional partial differential equations using deep learning

Jiequn Han et al.

PROCEEDINGS OF THE NATIONAL ACADEMY OF SCIENCES OF THE UNITED STATES OF AMERICA (2018)

Add to Collection

Article Multidisciplinary Sciences

Towards exact molecular dynamics simulations with machine-learned force fields

Stefan Chmiela et al.

NATURE COMMUNICATIONS (2018)

Add to Collection

Article Chemistry, Physical

Escaping Atom Types in Force Fields Using Direct Chemical Perception

David L. Mobley et al.

JOURNAL OF CHEMICAL THEORY AND COMPUTATION (2018)

Add to Collection

Article Biochemical Research Methods

CHARMM36m: an improved force field for folded and intrinsically disordered proteins

Jing Huang et al.

NATURE METHODS (2017)

Add to Collection

Article Chemistry, Multidisciplinary

ANI-1: an extensible neural network potential with DFT accuracy at force field computational cost

J. S. Smith et al.

CHEMICAL SCIENCE (2017)

Add to Collection

Review Chemistry, Multidisciplinary

An Empirical Polarizable Force Field Based on the Classical Drude Oscillator Model: Development History and Recent Applications

Justin A. Lemkul et al.

CHEMICAL REVIEWS (2016)

Add to Collection

Article Physics, Multidisciplinary

Molecular Dynamics with On-the-Fly Machine Learning of Quantum-Mechanical Forces

Zhenwei Li et al.

PHYSICAL REVIEW LETTERS (2015)

Add to Collection

Article Chemistry, Physical

Building Force Fields: An Automatic, Systematic, and Reproducible Approach

Lee-Ping Wang et al.

JOURNAL OF PHYSICAL CHEMISTRY LETTERS (2014)

Add to Collection

Article Computer Science, Hardware & Architecture

Data Science and Prediction

Vasant Dhar

COMMUNICATIONS OF THE ACM (2013)

Add to Collection

Article Chemistry, Physical

Calculating the Sensitivity and Robustness of Binding Free Energy Calculations to Force Field Parameters

Gabriel J. Rocklin et al.

JOURNAL OF CHEMICAL THEORY AND COMPUTATION (2013)

Add to Collection

Article Chemistry, Medicinal

Automation of the CHARMM General Force Field (CGenFF) II: Assignment of Bonded Parameters and Partial Atomic Charges

K. Vanommeslaeghe et al.

JOURNAL OF CHEMICAL INFORMATION AND MODELING (2012)

Add to Collection

Article Chemistry, Medicinal

Automation of the CHARMM General Force Field (CGenFF) I: Bond Perception and Atom Typing

K. Vanommeslaeghe et al.

JOURNAL OF CHEMICAL INFORMATION AND MODELING (2012)

Add to Collection

Article Chemistry, Physical

Force Field Benchmark of Organic Liquids: Density, Enthalpy of Vaporization, Heat Capacities, Surface Tension, Isothermal Compressibility, Volumetric Expansion Coefficient, and Dielectric Constant

Carl Caleman et al.

JOURNAL OF CHEMICAL THEORY AND COMPUTATION (2012)

Add to Collection

Editorial Material Engineering, Electrical & Electronic

The Fourth Paradigm: Data-Intensive Scientific Discovery

Kristin M. Tolle et al.

PROCEEDINGS OF THE IEEE (2011)

Add to Collection

Article Biophysics

Are current molecular dynamics force fields too helical?

Robert B. Best et al.

BIOPHYSICAL JOURNAL (2008)

Add to Collection

Article Physics, Multidisciplinary

Generalized neural-network representation of high-dimensional potential-energy surfaces

Joerg Behler et al.

PHYSICAL REVIEW LETTERS (2007)

Add to Collection

Review Chemistry, Multidisciplinary

Empirical force fields for biological macromolecules: Overview and issues

AD Mackerell

JOURNAL OF COMPUTATIONAL CHEMISTRY (2004)

Add to Collection

Article Computer Science, Artificial Intelligence

Ensembling neural networks: Many could be better than all

ZH Zhou et al.

ARTIFICIAL INTELLIGENCE (2002)

Add to Collection

Article Biochemistry & Molecular Biology

The Protein Data Bank

HM Berman et al.

NUCLEIC ACIDS RESEARCH (2000)

Add to Collection

Data science techniques in biomolecular force field development

Related references

Export Citation

Share Paper