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 subcellular localization using sequence-biased recurrent networks.

Mikael Bodén1, John Hawkins

  • 1School of Information Technology and Electrical Engineering, The University of Queensland, QLD 4072, Australia. mikael@itee.uq.edu.au

Bioinformatics (Oxford, England)
|March 5, 2005
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

TRIAGE Toolkit: Streamlined Discovery of Regulatory Genes and Elements.

Current protocols·2026
Same author

Improving Statin Prescribing for People Living With HIV: A Quality Improvement Initiative.

JACC. Case reports·2026
Same author

Ancestral Protein Reconstruction Uncovers a Thermotolerant Rieske Oxygenase with Enhanced <i>O</i>-Demethylation Activity toward 3-<i>O</i>-Methylgallate.

ACS synthetic biology·2026
Same author

Thermostable ancestors enabled evolutionary diversification of promiscuous chemical defence enzymes.

The EMBO journal·2026
Same author

A methionine-lined active site governs carbocation stabilization and product specificity in a bacterial terpene synthase.

FEBS letters·2026
Same author

Exploring the potential of naked barley to manage deoxynivalenol accumulation from Fusarium head blight.

TAG. Theoretical and applied genetics. Theoretische und angewandte Genetik·2026

Recurrent neural networks improve protein subcellular localization prediction accuracy. This computational method enhances the understanding of protein targeting signals for drug design and gene product annotation.

Area of Science:

  • Bioinformatics
  • Computational Biology
  • Molecular Biology

Background:

  • Protein targeting peptides direct nascent proteins to specific subcellular compartments.
  • Understanding protein localization is crucial for drug design and gene product annotation.
  • Computational prediction of protein subcellular localization is challenging due to sequence variability and dynamic sorting processes.

Purpose of the Study:

  • To develop and evaluate a novel sequence-biased recurrent network model for predicting protein subcellular localization.
  • To contrast the performance of recurrent networks with traditional feed-forward models like TargetP/SignalP.
  • To improve the accuracy and reliability of computational protein localization predictions.

Main Methods:

  • Implementation of a sequence-biased recurrent network architecture.

Related Experiment Videos

  • Evaluation of model performance at both residue and sequence levels.
  • Comparison with existing feed-forward network predictors (TargetP/SignalP).

    • Main Results:

    • Recurrent network models demonstrate improved prediction performance over feed-forward networks.
    • An ensemble of recurrent network models increased prediction accuracy by 6% for non-plant and 5% for plant data compared to TargetP.
    • The Protein Prowler web server incorporates the developed recurrent network predictor.

    Conclusions:

    • Sequence-biased recurrent networks offer a significant advancement in predicting protein subcellular localization.
    • The enhanced accuracy of these models facilitates more reliable protein function annotation and aids in drug discovery.
    • The Protein Prowler tool provides a valuable resource for researchers studying protein targeting and localization.