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

Consistent probabilistic outputs for protein function prediction.

Guillaume Obozinski1, Gert Lanckriet, Charles Grant

  • 1Department of Statistics, University of California, Berkeley, Berkeley, CA 94720, USA.

Genome Biology
|July 22, 2008
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

Prioritizing peptides for targeted mass spectrometry experiments using deep learning.

bioRxiv : the preprint server for biology·2026
Same author

Embryo-scale Visual Cell Sorting reveals a conserved transcriptomic signature of nucleolar size linked to proteostasis.

bioRxiv : the preprint server for biology·2026
Same author

Prediction and functional interpretation of inter-chromosomal genome architecture from DNA sequence with TwinC.

Nature communications·2026
Same author

Benchmarking Hi-C scaffolders using reference genomes and de novo assemblies.

Genome biology·2026
Same author

Unified imputation of missing data modalities and features in multi-omic data via shared representation learning.

bioRxiv : the preprint server for biology·2026
Same author

Accelerated first-order optimization under nonlinear constraints.

Mathematical programming·2026
Same journal

Integrated lipidomic and transcriptomic profiling of the host response in human malaria.

Genome biology·2026
Same journal

Centromeric satellite expansion drives genome evolution in the snowy owl.

Genome biology·2026
Same journal

Mapping the landscape of allele-specific expression in porcine genomes.

Genome biology·2026
Same journal

Genomic sequence evolution underlying human neocortical interareal diversification.

Genome biology·2026
Same journal

Regulatory mechanisms driven by functional 3'-UTR variants in alcohol use disorder and related traits.

Genome biology·2026
Same journal

A longitudinal single-nucleus transcriptomic atlas of bovine placentation reveals dynamic cellular hierarchies and regulatory programs.

Genome biology·2026
See all related articles

Predicting protein functions using Gene Ontology (GO) terms can lead to inconsistent results. This study introduces reconciliation methods to improve consistency and interpretability of GO term predictions.

Area of Science:

  • Bioinformatics
  • Computational Biology
  • Genomics

Background:

  • Hierarchical protein function prediction, like Gene Ontology (GO) term assignment, often uses independent predictions per term.
  • This independent approach can result in biologically inconsistent annotations (e.g., predicting a child term but not its parent).
  • Such inconsistencies hinder the interpretability of protein function predictions.

Purpose of the Study:

  • To develop and evaluate methods for reconciling independent protein function predictions.
  • To ensure predicted GO terms are consistent with the ontology's hierarchical structure.
  • To assess the impact of reconciliation on prediction accuracy (precision and recall).

Main Methods:

  • A baseline ensemble classifier was used for initial GO term prediction.

Related Experiment Videos

  • Eleven reconciliation methods were tested, including heuristic approaches, Bayesian networks, logistic regression extensions, and projection methods (isotonic regression, Kullback-Leibler projection).
  • Methods were evaluated across three prediction tasks: per term, per protein, and joint prediction.

    • Main Results:

    • The baseline prediction method produced frequent inconsistencies with the GO hierarchy.
    • Many reconciliation methods reduced prediction precision compared to unreconciled estimates.
    • Isotonic regression generally improved performance over the baseline and leveraged GO network constraints effectively.
    • Kullback-Leibler projection showed superior performance in high-precision joint evaluation scenarios.

    Conclusions:

    • Reconciliation is crucial for generating biologically consistent and interpretable protein function predictions.
    • Isotonic regression offers a robust method for improving GO term prediction accuracy and consistency.
    • Specific projection methods, like KL projection, excel in certain prediction tasks, highlighting the need for task-specific method selection.