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 Experiment Videos

Predicting enzyme class from protein structure without alignments.

Paul D Dobson1, Andrew J Doig

  • 1Department of Biomolecular Sciences, UMIST, P.O. Box 88, Manchester M60 1QD, UK.

Journal of Molecular Biology
|November 30, 2004
PubMed
Summary
This summary is machine-generated.

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

A machine learning classifier to identify and prioritise genes associated with murine cardiac development.

PLoS genetics·2026
Same author

Genetics can give evolving societies a leg up.

Nature·2025
Same author

Accurate and robust prediction of Amyloid-β brain deposition from plasma biomarkers and clinical information using machine learning.

Frontiers in aging neuroscience·2025
Same author

Essentiality, protein-protein interactions and evolutionary properties are key predictors for identifying cancer-associated genes using machine learning.

Scientific reports·2024
Same author

Predicting congenital renal tract malformation genes using machine learning.

Scientific reports·2023
Same author

Deconvolution of conformational exchange from Raman spectra of aqueous RNA nucleosides.

Communications chemistry·2023

This study introduces a novel method for predicting protein function directly from structural attributes, bypassing the need for sequence similarity searches. This approach accurately assigns enzyme classification numbers, aiding in the annotation of unclassified proteins.

Area of Science:

  • Structural bioinformatics
  • Computational biology
  • Protein science

Background:

  • Increasingly, protein structures are determined faster than their functions can be experimentally annotated.
  • Current protein function prediction methods often rely on identifying homologous proteins, which fails when no similar proteins are found or if they also lack annotations.

Purpose of the Study:

  • To develop and validate a method for predicting protein function directly from structural features, independent of sequence similarity.
  • To assess the utility of simple structural attributes for assigning Enzyme Classification (EC) numbers.
  • To apply the developed method to unclassified proteins in the Protein Data Bank.

Main Methods:

  • Extraction of simple structural attributes (secondary structure content, amino acid propensities, surface properties, ligands) from crystal structures.

Related Experiment Videos

  • Classification of enzymes based on Enzyme Classification (EC) numbers.
  • Implementation of one-class versus one-class support vector machine models for function prediction.

    • Main Results:

    • Achieved 35% accuracy with the top-ranked prediction and 60% accuracy with the top two ranks for assigning EC numbers.
    • Demonstrated the effectiveness of using basic structural attributes for protein function prediction.
    • Successfully applied the method to predict functions for currently unclassified proteins in the Protein Data Bank.

    Conclusions:

    • Simple structural attributes are valuable for predicting protein function, offering an alternative to alignment-based methods.
    • The developed method provides a robust approach for annotating protein function, particularly for proteins lacking clear homologs.
    • This work highlights the strong correlation between protein structure and function.