Search research articles

ABOUT JoVE

Overview Leadership Blog JoVE Help Center

AUTHORS

Publishing Process Editorial Board Scope & Policies Peer Review FAQ Submit

LIBRARIANS

Testimonials Subscriptions Access Resources Library Advisory Board FAQ

RESEARCH

JoVE Journal Methods Collections JoVE Encyclopedia of Experiments Archive

EDUCATION

JoVE Core JoVE Business JoVE Science Education JoVE Lab Manual Faculty Resource Center Faculty Site

Terms & Conditions of Use

Related Concept Videos

Predicting Molecular Geometry

Predicting Molecular Geometry

VSEPR Theory for Determination of Electron Pair Geometries

The Energies of Atomic Orbitals

The Energies of Atomic Orbitals

In an atom, the negatively charged electrons are attracted to the positively charged nucleus. In a multielectron atom, electron-electron repulsions are also observed. The attractive and repulsive forces are dependent on the distance between the particles, as well as the sign and magnitude of the charges on the individual particles. When the charges on the particles are opposite, they attract each other. If both particles have the same charge, they repel each other.

Vector Representation of Complex Numbers

Vector Representation of Complex Numbers

Complex numbers, represented in Cartesian coordinates, can also be visualized as vectors. These vectors can be expressed in polar form, emphasizing their magnitude and angle. When a complex number is input into a function, the output is another complex number, highlighting the function's zero point from which the vector representation can originate.
Consider a function defined as the product of the complex factors in the numerator divided by the product of the complex factors in the...

Molecular Geometry and Dipole Moments

Molecular Geometry and Dipole Moments

The VSEPR theory can be used to determine the electron pair geometries and molecular structures as follows:

Force Vector along a Line

Force Vector along a Line

Quite often in three-dimensional statics problems, the direction of a force is specified by two points through which its line of action passes. Consider a three-dimensional static pole with a cable anchored to the ground.

Kinetic Molecular Theory: Molecular Velocities, Temperature, and Kinetic Energy

Kinetic Molecular Theory: Molecular Velocities, Temperature, and Kinetic Energy

The kinetic molecular theory qualitatively explains the behaviors described by the various gas laws. The postulates of this theory may be applied in a more quantitative fashion to derive these individual laws.

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by

Same author

Unifying p<i>K</i><sub>a</sub> and Protonation Prediction with Sequence-Based Deep Learning.

Journal of chemical information and modeling·2026

Same author

End-to-end molecular structure elucidation from multimodal NMR spectra images using vision transformers.

Chemical science·2026

Same author

A Novel Covalent Inhibitor Fragment for the SARS-CoV-2 Main Protease Identified by Target-Specific Deep Learning.

ACS chemical biology·2026

Same author

ProMol_Func: A Structure-Free Deep Learning Model for Virtual Screening.

JACS Au·2026

Same author

Fine-Tuning DiffDock-L for Allosteric Kinase Docking.

Journal of chemical information and modeling·2026

Same author

Correction: Structure-based design of an aromatic helical foldamer-protein interface.

Chemical science·2026

Search research articles

Related Experiment Video

Updated: Jan 24, 2026

Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications

Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications

Published on: December 15, 2023

Predicting Molecular Energy Using Force-Field Optimized Geometries and Atomic Vector Representations Learned from an

Jianing Lu¹, Cheng Wang¹, Yingkai Zhang^1,2

¹Department of Chemistry , New York University , New York , New York 10003 , United States.

Journal of Chemical Theory and Computation

|May 31, 2019

Summary

Machine learning models can now predict molecular energies using force-field optimized geometries, bypassing computationally expensive DFT calculations. This approach achieves high accuracy, enabling efficient molecular conformation analysis.

More Related Videos

Deep Neural Networks for Image-Based Dietary Assessment

Deep Neural Networks for Image-Based Dietary Assessment

Published on: March 13, 2021

A Step-by-Step Implementation of DeepBehavior, Deep Learning Toolbox for Automated Behavior Analysis

A Step-by-Step Implementation of DeepBehavior, Deep Learning Toolbox for Automated Behavior Analysis

Published on: February 6, 2020

Related Experiment Videos

Last Updated: Jan 24, 2026

Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications

Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications

Published on: December 15, 2023

Deep Neural Networks for Image-Based Dietary Assessment

Deep Neural Networks for Image-Based Dietary Assessment

Published on: March 13, 2021

A Step-by-Step Implementation of DeepBehavior, Deep Learning Toolbox for Automated Behavior Analysis

A Step-by-Step Implementation of DeepBehavior, Deep Learning Toolbox for Automated Behavior Analysis

Published on: February 6, 2020

Area of Science:

Computational Chemistry
Materials Science
Machine Learning

Background:

Predicting molecular properties often requires computationally intensive quantum mechanical (QM) calculations.
Existing machine learning models typically rely on geometries optimized using Density Functional Theory (DFT), limiting practical applications.
Deep Tensor Neural Networks (DTNN) have shown promise but often require DFT-optimized inputs.

Purpose of the Study:

To investigate the use of machine learning for predicting molecular atomization energies and conformation stability using geometries optimized with the Merck Molecular Force Field (MMFF).
To adapt and improve the DTNN approach for efficient training and hyperparameter optimization.
To develop transfer learning strategies for accurate molecular energy prediction using MMFF geometries.

Main Methods:

Developed an improved Deep Tensor Neural Network (DTNN_7ib) model with enhanced training efficiency and extensive hyperparameter search.
Applied a transfer learning (TL) strategy using atomic vector representations from DTNN_7ib to train models on datasets with MMFF-optimized geometries (QM9_M and eMol9_C_M).
Evaluated model performance using Mean Absolute Error (MAE) on QM9, QM9_M, and eMol9_C_M datasets.

Main Results:

The DTNN_7ib model achieved a test accuracy of 0.34 kcal/mol MAE on the QM9 dataset.
The TL_QM9_M model, using MMFF geometries, reached an MAE of 0.79 kcal/mol.
A transfer learning model for the eMol9_C_M dataset achieved an MAE of 0.51 kcal/mol using MMFF geometries.

Conclusions:

Machine learning models can achieve DFT-level accuracy for molecular energy prediction using computationally cheaper, force-field optimized geometries.
The integration of molecular modeling and machine learning offers a powerful approach for advanced molecular conformation analysis.
This work demonstrates the potential of DTNN and transfer learning for efficient and accurate computational chemistry applications.