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 Reaction Outcomes02:24

Predicting Reaction Outcomes

8.7K
Kinetics describes the rate and path by which a reaction occurs. In contrast, thermodynamics deals with state functions and describes the properties, behavior, and components of a system. It is not concerned with the path taken by the process and cannot address the rate at which a reaction occurs. Although it does provide information about what can happen during a reaction process, it does not describe the detailed steps of what appears on an atomic or a molecular level. On the other hand,...
8.7K
Survival Tree01:19

Survival Tree

173
Survival trees are a non-parametric method used in survival analysis to model the relationship between a set of covariates and the time until an event of interest occurs, often referred to as the "time-to-event" or "survival time." This method is particularly useful when dealing with censored data, where the event has not occurred for some individuals by the end of the study period, or when the exact time of the event is unknown.
 Building a Survival Tree
Constructing a...
173
Classification of Titrimetric Analysis Based on Reaction Types01:01

Classification of Titrimetric Analysis Based on Reaction Types

888
Titrimetric analysis in solution chemistry involves measuring the volume of solutions and is often called volumetric analysis. The standard solution of known concentration in the burette is called the titrant, whereas the solution of unknown concentration in the flask is called the analyte, or titrand. Titrimetric analyses can be classified into four types based on the reactions between the titrant and analyte.
Titrations between an acid and a base lead to neutralization reactions that form...
888
Inductive Effects on Chemical Shift: Overview01:27

Inductive Effects on Chemical Shift: Overview

1.4K
The protons in unsubstituted alkanes are strongly shielded with chemical shifts below 1.8 ppm. Methine, methylene, and methyl protons appear at approximately 1.7, 1.2 and 0.7 ppm, while the proton signal from methane appears at 0.23 ppm. An electronegative substituent, such as chlorine, withdraws the electron density from the protons, increasing their chemical shift. Progressive substitution of the hydrogens in methane by chlorine shifts the proton signals increasingly downfield, to 3.05 ppm in...
1.4K
Improving Translational Accuracy02:07

Improving Translational Accuracy

12.0K
Base complementarity between the three base pairs of mRNA codon and the tRNA anticodon is not a failsafe mechanism. Inaccuracies can range from a single mismatch to no correct base pairing at all. The free energy difference between the correct and nearly correct base pairs can be as small as 3 kcal/ mol. With complementarity being the only proofreading step, the estimated error frequency would be one wrong amino acid in every 100 amino acids incorporated. However, error frequencies observed in...
12.0K
Predicting Products: SN1 vs. SN202:27

Predicting Products: SN1 vs. SN2

14.2K
Nucleophilic substitution reactions of alkyl halides can proceed via an SN1 or an SN2 mechanism. While in SN2 reactions, the nucleophile attacks the substrate simultaneously as the leaving group departs, in SN1 reactions, the substrate first dissociates to give the carbocation intermediate. Various factors such as the structure of the substrate, the strength of the nucleophile, and the nature of the solvent promote one mechanism over the other.
With increased substitution on the alkyl halide,...
14.2K

You might also read

Related Articles

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

Sort by
Same author

Machine-Learned Leftmost Hessian Eigenvectors for Robust Transition State Finding.

Journal of chemical theory and computation·2026
Same author

Ion correlations explain kinetic selectivity in diffusion-limited solid-state synthesis reactions.

Nature materials·2026
Same author

Two-Dimensional Mesoporous Tungsten Oxynitride/Carbon Nanosheets: Enabling High-Performance Lithium-Sulfur Batteries.

Angewandte Chemie (International ed. in English)·2026
Same author

Identification of Solid-Electrolyte Interphase Species by Joint Characterization of Li-Ion Battery Chemistry by Mass Spectrometry and Electrochemical Reaction Networks.

Journal of the American Chemical Society·2026
Same author

Foundation models for atomistic simulation of chemistry and materials.

Nature reviews. Chemistry·2026
Same author

Accelerated Discovery of Cost-Effective Photoabsorber Materials for Near-Infrared (λ = 1600 nm) Photodetector Applications.

ACS materials Au·2026

Related Experiment Video

Updated: Oct 2, 2025

Integration of Animal Behavioral Assessment and Convolutional Neural Network to Study Wasabi-Alcohol Taste-Smell Interaction
06:19

Integration of Animal Behavioral Assessment and Convolutional Neural Network to Study Wasabi-Alcohol Taste-Smell Interaction

Published on: August 16, 2024

575

Improving machine learning performance on small chemical reaction data with unsupervised contrastive pretraining.

Mingjian Wen1, Samuel M Blau1, Xiaowei Xie2,3

  • 1Energy Technologies Area, Lawrence Berkeley National Laboratory Berkeley CA 94720 USA.

Chemical Science
|February 28, 2022
PubMed
Summary

Leveraging unlabeled chemical reaction data with contrastive learning significantly improves machine learning model accuracy and transferability, especially for small labeled datasets. This approach enhances reaction classification and navigates chemical space more effectively than traditional methods.

Related Experiment Videos

Last Updated: Oct 2, 2025

Integration of Animal Behavioral Assessment and Convolutional Neural Network to Study Wasabi-Alcohol Taste-Smell Interaction
06:19

Integration of Animal Behavioral Assessment and Convolutional Neural Network to Study Wasabi-Alcohol Taste-Smell Interaction

Published on: August 16, 2024

575

Area of Science:

  • Computational Chemistry
  • Machine Learning in Chemistry
  • Chemical Informatics

Background:

  • Machine learning (ML) models, particularly graph neural networks (GNNs), require extensive labeled data for accurate chemical reaction prediction and analysis.
  • Overfitting and poor transferability are significant challenges when training ML models on limited labeled chemical reaction datasets.
  • Traditional methods like rule-driven fingerprints struggle with small datasets and complex chemical insights.

Purpose of the Study:

  • To develop a strategy for training accurate ML models for chemical reactions using limited labeled data by leveraging abundant unlabeled data.
  • To improve the accuracy and transferability of GNN-based models for chemical reaction classification.
  • To demonstrate the utility of learned representations for reaction similarity searching and other predictive tasks.

Main Methods:

  • Utilized unsupervised contrastive learning on unlabeled reaction data to pretrain a GNN model.
  • Developed chemically consistent reaction augmentation techniques that preserve the reaction center for effective representation learning.
  • Fine-tuned the pretrained model on a small set of labeled reactions for the target task (reaction classification).

Main Results:

  • The contrastive pretraining strategy significantly improved the performance of the GNN model compared to training from scratch on limited labeled data.
  • The transfer-learned model outperformed traditional rule-driven reaction fingerprints and masked language model-derived fingerprints.
  • Learned GNN-based reaction fingerprints demonstrated effectiveness in reaction classification and navigating chemical reaction space via similarity queries.

Conclusions:

  • Unsupervised contrastive learning on unlabeled data is a powerful strategy to enhance ML models for chemical reaction prediction with limited labels.
  • Chemically informed augmentation is crucial for extracting relevant information from unlabeled data.
  • This approach offers a viable solution for building accurate and transferable ML models in data-scarce chemical discovery scenarios.