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

Predicting Molecular Geometry02:27

Predicting Molecular Geometry

46.4K
VSEPR Theory for Determination of Electron Pair Geometries
46.4K
Classification of Neurotransmitters01:30

Classification of Neurotransmitters

5.6K
Neurotransmitters play a crucial role in the communication between neurons in the autonomic nervous system. Neurons in the autonomic nervous system can be cholinergic or adrenergic depending on the neurotransmitters synthesized. Cholinergic neurons use acetylcholine as their primary neurotransmitter. This includes all the preganglionic fibers of the sympathetic and pre- and postganglionic fibers of the parasympathetic nervous systems. In addition, neurons of the somatic nervous system also use...
5.6K
Molecular Models02:00

Molecular Models

44.2K
Physical models representing molecular architectures of chemical compounds play essential roles in understanding chemistry. The use of molecular models makes it easier to visualize the structures and shapes of atoms and molecules.
44.2K
Hückel's Rule Diagram of π MOs: Frost Circle01:08

Hückel's Rule Diagram of π MOs: Frost Circle

6.0K
The Frost circle or the inscribed polygon method is a graphical method for determining the relative energies of π molecular orbitals (MOs) for planar, fully conjugated, and monocyclic compounds. This method was first described by A. A. Frost and Boris Musulin in 1953.
A Frost circle is constructed by drawing a polygon whose number of edges is equal to the number of carbons of the given cyclic system, with one of the vertices pointing down. Then, a circle is drawn enclosing the polygon so that...
6.0K
Fermi Level01:18

Fermi Level

2.0K
The Fermi-Dirac function is represented by an S-shaped curve indicating the probability of an energy state being occupied by an electron at a given temperature. The Fermi level is the energy level at which there is a fifty percent chance of finding an electron, and it is positioned between the lower-energy valence band and the higher-energy conduction band.
At absolute zero temperature, electrons fill all energy states up to the Fermi level, leaving upper states empty. As the temperature rises,...
2.0K
Fermi Level Dynamics01:12

Fermi Level Dynamics

838
The vacuum level denotes the energy threshold required for an electron to escape from a material surface. It is usually positioned above the conduction band of a semiconductor and acts as a benchmark for comparing electron energies within various materials.
Electron affinity in semiconductors refers to the energy gap between the minimum of its conduction band and the vacuum level and it is a critical parameter in determining how easily a semiconductor can accept additional electrons.
The work...
838

You might also read

Related Articles

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

Sort by
Same author

Molecular Modeling of Surfactant Interaction on Phospholipid Bilayers Mimicking Corneal Epithelium.

Journal of chemical theory and computation·2026
Same author

Out-of-distribution evaluation of active learning pipelines for molecular property prediction.

RSC advances·2026
Same author

Unravelling the dynamics of the maturation protein in MS2 bacteriophage via molecular simulations.

Scientific reports·2025
Same author

Designing Moderately-Solvating Electrolytes for High-Performance Lithium-Sulfur Batteries.

Advanced materials (Deerfield Beach, Fla.)·2025
Same author

Extracting Material Property Measurements from Scientific Literature with Limited Annotations.

Journal of chemical information and modeling·2025
Same author

ezAlign: A Tool for Converting Coarse-Grained Molecular Dynamics Structures to Atomistic Resolution for Multiscale Modeling.

Molecules (Basel, Switzerland)·2024

Related Experiment Video

Updated: Feb 26, 2026

Using Three-color Single-molecule FRET to Study the Correlation of Protein Interactions
11:22

Using Three-color Single-molecule FRET to Study the Correlation of Protein Interactions

Published on: January 30, 2018

10.7K

FragNet: A Graph Neural Network for Molecular Property Prediction with Four Levels of Interpretability.

Gihan Panapitiya1, Peiyuan Gao1, C Mark Maupin1

  • 1Pacific Northwest National Laboratory, Richland, Washington 99354, United States.

Journal of the American Chemical Society
|February 25, 2026
PubMed
Summary

This study introduces an interpretable graph neural network for molecular property prediction. The model achieves high accuracy and reveals which molecular components significantly influence predicted properties, aiding scientific discovery.

More Related Videos

Evidence-based Knowledge Synthesis and Hypothesis Validation: Navigating Biomedical Knowledge Bases via Explainable AI and Agentic Systems
05:47

Evidence-based Knowledge Synthesis and Hypothesis Validation: Navigating Biomedical Knowledge Bases via Explainable AI and Agentic Systems

Published on: June 13, 2025

1.7K
Modeling the Functional Network for Spatial Navigation in the Human Brain
05:55

Modeling the Functional Network for Spatial Navigation in the Human Brain

Published on: October 13, 2023

1.6K

Related Experiment Videos

Last Updated: Feb 26, 2026

Using Three-color Single-molecule FRET to Study the Correlation of Protein Interactions
11:22

Using Three-color Single-molecule FRET to Study the Correlation of Protein Interactions

Published on: January 30, 2018

10.7K
Evidence-based Knowledge Synthesis and Hypothesis Validation: Navigating Biomedical Knowledge Bases via Explainable AI and Agentic Systems
05:47

Evidence-based Knowledge Synthesis and Hypothesis Validation: Navigating Biomedical Knowledge Bases via Explainable AI and Agentic Systems

Published on: June 13, 2025

1.7K
Modeling the Functional Network for Spatial Navigation in the Human Brain
05:55

Modeling the Functional Network for Spatial Navigation in the Human Brain

Published on: October 13, 2023

1.6K

Area of Science:

  • Computational chemistry
  • Materials science
  • Drug discovery

Background:

  • Accurate molecular property prediction is crucial for fields like drug development and materials design.
  • Existing machine learning models often lack interpretability, hindering scientific insight.
  • There is a need for models that provide both high accuracy and transparency in predictions.

Purpose of the Study:

  • To develop a graph neural network (GNN) model for molecular property prediction.
  • To enhance model interpretability by analyzing contributions from molecular substructures.
  • To provide insights into the relationships between molecular structure and predicted properties.

Main Methods:

  • Development of a novel graph neural network architecture.
  • Incorporation of multi-level substructure analysis (atoms, bonds, fragments, connections).
  • Quantification of the impact of individual fragments on property predictions.

Main Results:

  • The GNN model achieves prediction accuracies comparable to leading state-of-the-art models.
  • The model successfully identifies significant atoms, bonds, fragments, and their connections.
  • It quantifies the influence of specific fragments on property values, enabling targeted optimization.

Conclusions:

  • The developed GNN offers a powerful tool for accurate and interpretable molecular property prediction.
  • Its ability to dissect molecular contributions facilitates deeper scientific understanding and accelerates material and drug design.
  • Interpretable features are key for deriving actionable insights from machine learning models in chemistry.