Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Concept Videos

You might also read

Related Articles

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

Sort by
Same author

Maximizing Room-Temperature Red Phosphorescence in Contorted Hexabenzocoronene Derivatives.

Chemistry of materials : a publication of the American Chemical Society·2026
Same author

Modulating the Rate of Charge Recombination in Bichromophore End-Capped Hairpin DNA Sequences.

The journal of physical chemistry letters·2026
Same author

Ion diffusion overestimates figures of merit in polymeric mixed conductors.

Nature communications·2026
Same author

Machine Learning for Materials Chemistry: New Frontiers and Emerging Paradigms.

ACS applied materials & interfaces·2025
Same author

Self-doped and planar nitrogen-bay-functionalized perylene diimides: unravelling structure-property relationships.

Chemical science·2025
Same author

A'-Site Dipole Magnitude and Direction Dominate the Ionization Energy and Electron Affinity of Layered Metal-Halide Perovskites.

Journal of the American Chemical Society·2025

Related Experiment Video

Updated: Apr 11, 2026

Automated Protocols for Macromolecular Crystallization at the MRC Laboratory of Molecular Biology
11:20

Automated Protocols for Macromolecular Crystallization at the MRC Laboratory of Molecular Biology

Published on: January 24, 2018

17.1K

Machine Learning Potential-Enabled Platform for the In Silico Design of Functional Organic Molecular Crystals.

Vinayak Bhat1,2, Chad Risko1

  • 1Department of Chemistry & Center of Applied Energy Research, University of Kentucky, Lexington, Kentucky 40506, United States.

Journal of Chemical Information and Modeling
|April 9, 2026
PubMed
Summary

We developed a machine learning (ML) potential for predicting molecular crystal properties, accelerating materials discovery. This ML potential, trained on a large dataset, accurately predicts energies and forces, enabling faster design of functional organic materials.

More Related Videos

Application of I TASSER, trRosetta, UCSF Chimera, HADDOCK server, and HEX loria for De Novo and In Silico Design of Proteins
05:08

Application of I TASSER, trRosetta, UCSF Chimera, HADDOCK server, and HEX loria for De Novo and In Silico Design of Proteins

Published on: July 8, 2025

1.3K
Protein WISDOM: A Workbench for In silico De novo Design of BioMolecules
10:58

Protein WISDOM: A Workbench for In silico De novo Design of BioMolecules

Published on: July 25, 2013

17.7K

Related Experiment Videos

Last Updated: Apr 11, 2026

Automated Protocols for Macromolecular Crystallization at the MRC Laboratory of Molecular Biology
11:20

Automated Protocols for Macromolecular Crystallization at the MRC Laboratory of Molecular Biology

Published on: January 24, 2018

17.1K
Application of I TASSER, trRosetta, UCSF Chimera, HADDOCK server, and HEX loria for De Novo and In Silico Design of Proteins
05:08

Application of I TASSER, trRosetta, UCSF Chimera, HADDOCK server, and HEX loria for De Novo and In Silico Design of Proteins

Published on: July 8, 2025

1.3K
Protein WISDOM: A Workbench for In silico De novo Design of BioMolecules
10:58

Protein WISDOM: A Workbench for In silico De novo Design of BioMolecules

Published on: July 25, 2013

17.7K

Area of Science:

  • Materials Science
  • Computational Chemistry
  • Crystallography

Background:

  • In silico approaches significantly accelerate the design and discovery of functional molecular materials.
  • Machine learning (ML) models are crucial for analyzing and manipulating molecular and crystal structures in materials design.
  • Accurate ML potentials are essential for predicting energies and forces on atoms within crystal structures.

Purpose of the Study:

  • To present a novel ML potential trained on the OCELOT Crystal Relaxation v1 dataset for predicting properties of molecular crystals.
  • To develop workflows for analyzing crystal surfaces, morphologies, and manipulating crystal structures using the ML potential.
  • To create an accessible web-based tool, OCELOT XtalTransform, for utilizing these ML capabilities.

Main Methods:

  • Developed an ML potential trained on 15,000 molecular crystal structures and over 2.2 million crystal geometries from the OCELOT Crystal Relaxation v1 dataset.
  • Validated the ML potential against density functional theory (DFT) calculations, achieving a mean absolute error of 0.008 eV/atom for energy and 0.034 eV/Å for forces.
  • Created integrated workflows for crystal surface analysis, morphology studies, and structure manipulation and relaxation.

Main Results:

  • The ML potential accurately predicts energies and forces for crystal structures of π-conjugated organic molecules.
  • Achieved high accuracy compared to DFT calculations, demonstrating the robustness of the developed ML potential.
  • Successfully deployed the ML potential and associated workflows via the OCELOT XtalTransform web tool.

Conclusions:

  • The developed ML potential and OCELOT XtalTransform tool significantly accelerate the in silico discovery of functional molecular crystals.
  • Democratizes access to advanced ML potentials for organic molecular crystals, fostering broader scientific community engagement.
  • Provides new avenues for materials design and discovery by bridging sophisticated simulations with user-friendly interfaces.