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

Introduction to Metabolism01:30

Introduction to Metabolism

Metabolism encompasses all biochemical reactions in a living organism, facilitating both the breakdown and synthesis of biomolecules. These metabolic processes are categorized into catabolic and anabolic pathways, which operate in a coordinated manner to ensure energy balance and cellular function.Catabolic Pathways and Energy ReleaseCatabolic pathways involve the breakdown of complex macromolecules such as carbohydrates, lipids, and proteins into smaller structures like monosaccharides, fatty...
Molecular Models02:00

Molecular Models

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.
Protein Networks02:26

Protein Networks

An organism can have thousands of different proteins, and these proteins must cooperate to ensure the health of an organism. Proteins bind to other proteins and form complexes to carry out their functions. Many proteins interact with multiple other proteins creating a complex network of protein interactions.
These interactions can be represented through maps depicting protein-protein interaction networks, represented as nodes and edges. Nodes are circles that are representative of a protein,...
Structure-Activity Relationships and Drug Design01:28

Structure-Activity Relationships and Drug Design

Drug design is a dynamic field that involves discovering and developing new medications based on specific biological targets. This process heavily relies on structure-activity relationships (SAR) and quantitative structure-activity relationships (QSAR) to guide the design and optimization of efficient drugs.
SAR studies the intricate relationship between a drug's chemical structure and biological activity. It focuses on understanding how modifications to a drug's structure can influence its...
Regulation of Metabolism01:19

Regulation of Metabolism

Cellular needs and conditions vary from cell to cell and change within individual cells over time. For example, the required enzymes and energetic demands of stomach cells are different from those of fat storage cells, skin cells, blood cells, and nerve cells. Furthermore, a digestive cell works much harder to process and break down nutrients during the time that closely follows a meal compared with many hours after a meal. As these cellular demands and conditions vary, so do the amounts and...

You might also read

Related Articles

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

Sort by
Same author

The recount3 Python package for programmatic access to uniformly processed RNA-seq data.

bioRxiv : the preprint server for biology·2026
Same author

Information-Content-Informed Kendall-Tau Correlation Methodology: Interpreting Missing Values in Metabolomics as Potentially Useful Information.

Metabolites·2026
Same author

FASN Inhibition Enhances the Efficacy of Chemotherapy in Colorectal Cancer by Inhibiting the DNA Damage Response.

Cancer research·2026
Same author

A Major Update and Improved Validation Functionality in the mwtab Python Library and the Metabolomics Workbench File Status Website.

Metabolites·2026
Same author

Major Update and Improved Validation Functionality in the mwtab Python Library and the Metabolomics Workbench File Status Website.

bioRxiv : the preprint server for biology·2026
Same author

Predicting the pathway involvement of metabolites annotated in the MetaCyc knowledgebase.

BMC bioinformatics·2026

Related Experiment Video

Updated: Jun 26, 2026

A Web Tool for Generating High Quality Machine-readable Biological Pathways
08:01

A Web Tool for Generating High Quality Machine-readable Biological Pathways

Published on: February 8, 2017

Chemical Structure Representation Standardization Is Needed to Generalize Metabolite-Pathway Involvement Prediction

Erik D Huckvale1, Hunter N B Moseley1,2,3,4,5

  • 1Markey Cancer Center, University of Kentucky, Lexington, KY 40536, USA.

Metabolites
|June 25, 2026
PubMed
Summary

This study introduces a large machine learning dataset for predicting metabolite pathway involvement, improving accuracy by standardizing chemical structures. Standardized representations are crucial for robust predictions in novel chemical structures.

Keywords:
InChI canonicalizationKEGGMetaCycReactomecompoundsdata curationmachine learningmodel generalizabilityneural networkspathways

More Related Videos

Annotation of Plant Gene Function via Combined Genomics, Metabolomics and Informatics
08:09

Annotation of Plant Gene Function via Combined Genomics, Metabolomics and Informatics

Published on: June 17, 2012

CorrelationCalculator and Filigree: Tools for Data-Driven Network Analysis of Metabolomics Data
07:11

CorrelationCalculator and Filigree: Tools for Data-Driven Network Analysis of Metabolomics Data

Published on: November 10, 2023

Related Experiment Videos

Last Updated: Jun 26, 2026

A Web Tool for Generating High Quality Machine-readable Biological Pathways
08:01

A Web Tool for Generating High Quality Machine-readable Biological Pathways

Published on: February 8, 2017

Annotation of Plant Gene Function via Combined Genomics, Metabolomics and Informatics
08:09

Annotation of Plant Gene Function via Combined Genomics, Metabolomics and Informatics

Published on: June 17, 2012

CorrelationCalculator and Filigree: Tools for Data-Driven Network Analysis of Metabolomics Data
07:11

CorrelationCalculator and Filigree: Tools for Data-Driven Network Analysis of Metabolomics Data

Published on: November 10, 2023

Area of Science:

  • Bioinformatics
  • Computational Biology
  • Cheminformatics

Background:

  • Metabolite pathway annotations are crucial for biological experiments but are often incomplete in knowledgebases like KEGG, Reactome, and MetaCyc.
  • Experimental or manual curation of these annotations is costly and time-consuming.
  • Machine learning (ML) offers a solution, but models trained on multiple knowledgebases face challenges due to inconsistent chemical structure representations.

Purpose of the Study:

  • To develop the largest machine learning-ready dataset for compound-pathway involvement prediction.
  • To train and evaluate a robust ML model capable of predicting a high number of pathway annotations.
  • To investigate the impact of standardized chemical structure representations on model performance and generalizability.

Main Methods:

  • Constructed a dataset of approximately 50 million compound-pathway annotations from KEGG, Reactome, and MetaCyc.
  • Employed a multitask classification, graph convolutional neural network-like model using chemical substructure features.
  • Standardized chemical structure representations using InChI (IUPAC International Chemical Identifier) canonicalization to address inconsistencies.

Main Results:

  • The standardized dataset achieved a higher mean Matthews correlation coefficient (MCC) of 0.9036 ± 0.0033 compared to the non-standardized dataset's 0.8725 ± 0.0064.
  • Model generalizability significantly improved with standardized representations, showing only a 0.0384 drop in mean MCC versus a 0.2687 drop for non-standardized data.
  • The developed model predicts the highest number of pathway annotations to date.

Conclusions:

  • A comprehensive ML dataset for compound-pathway prediction has been created.
  • Standardizing chemical structure representation is essential for building accurate and generalizable ML models in this domain.
  • This approach enhances the prediction of novel compound-pathway associations.