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

Prediction of protein solvent accessibility using support vector machines.

Zheng Yuan1, Kevin Burrage, John S Mattick

  • 1Institute for Molecular Bioscience and ARC Special Centre for Functional and Applied Genomics, The University of Queensland, Brisbane, Australia. z.yuan@imb.uq.edu.au

Proteins
|July 12, 2002
PubMed
Summary

A Support Vector Machine (SVM) model accurately predicts protein solvent accessibility from primary sequences. This machine learning approach offers improved accuracy compared to other methods for biological sequence analysis.

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

FIRST-seq: a nanopore-based cDNA sequencing platform for RNA modification and structure profiling.

Genome biology·2026
Same author

A mathematical perspective on hypothesis-driven model construction: A case study in pea.

Mathematical biosciences·2026
Same author

Specificity of the stabilizing interaction between intrinsically disordered protein sequences and G-quadruplexes in RNA.

Nucleic acids research·2026
Same author

The emerging roles of long non-coding RNAs in the nervous system.

Nature reviews. Neuroscience·2025
Same author

ECSFinder: optimized prediction of evolutionarily conserved RNA secondary structures from genome sequences.

Nucleic acids research·2025
Same author

Optimal control of multiple myeloma assuming drug resistance and off-target effects.

PLoS computational biology·2025

Area of Science:

  • Computational Biology
  • Bioinformatics
  • Machine Learning

Background:

  • Protein solvent accessibility is crucial for understanding protein structure and function.
  • Predicting solvent accessibility from primary sequences remains a challenge in bioinformatics.

Purpose of the Study:

  • To develop and evaluate a Support Vector Machine (SVM) learning system for predicting protein solvent accessibility.
  • To explore the impact of different kernel functions and sliding window sizes on prediction performance.

Main Methods:

  • Trained a Support Vector Machine (SVM) model using primary protein sequence data.
  • Investigated various kernel functions and sliding window sizes.
  • Utilized a 15% cut-off threshold for binary classification of solvent accessibility (exposed/buried).

Related Experiment Videos

Main Results:

  • Achieved 70.1% accuracy for single sequence input and 73.9% for multiple alignment sequence input.
  • SVM predictions for three or more states of solvent accessibility were comparable or superior to existing methods (e.g., neural networks, Bayesian classification).

Conclusions:

  • The Support Vector Machine (SVM) method is a highly effective tool for biological sequence analysis, specifically for predicting protein solvent accessibility.
  • The SVM system shows potential for integration with other prediction methods to enhance reliability.